Loading [Contrib]/a11y/accessibility-menu.js
Research article Special Issues

A convolutional neural network-based linguistic steganalysis for synonym substitution steganography

  • Received: 25 January 2019 Accepted: 24 October 2019 Published: 11 November 2019
  • In this paper, a linguistic steganalysis method based on two-level cascaded convolutional neural networks (CNNs) is proposed to improve the system's ability to detect stego texts, which are generated via synonym substitutions. The first-level network, sentence-level CNN, consists of one convolutional layer with multiple convolutional kernels in different window sizes, one pooling layer to deal with variable sentence lengths, and one fully connected layer with dropout as well as a softmax output, such that two final steganographic features are obtained for each sentence. The unmodified and modified sentences, along with their words, are represented in the form of pre-trained dense word embeddings, which serve as the input of the network. Sentence-level CNN provides the representation of a sentence, and can thus be utilized to predict whether a sentence is unmodified or has been modified by synonym substitutions. In the second level, a text-level CNN exploits the predicted representations of sentences obtained from the sentence-level CNN to determine whether the detected text is a stego text or cover text. Experimental results indicate that the proposed sentence-level CNN can effectively extract sentence features for sentence-level steganalysis tasks and reaches an average accuracy of 82.245%. Moreover, the proposed steganalysis method achieves greatly improved detection performance when distinguishing stego texts from cover texts.

    Citation: Lingyun Xiang, Guoqing Guo, Jingming Yu, Victor S. Sheng, Peng Yang. A convolutional neural network-based linguistic steganalysis for synonym substitution steganography[J]. Mathematical Biosciences and Engineering, 2020, 17(2): 1041-1058. doi: 10.3934/mbe.2020055

    Related Papers:

    [1] Kazuo Sano . Intelligence and global bias in the stock market. Data Science in Finance and Economics, 2023, 3(2): 184-195. doi: 10.3934/DSFE.2023011
    [2] László Bokor . Alarm probabilities of simple tests of merger control: An analytic derivation. Data Science in Finance and Economics, 2024, 4(2): 188-217. doi: 10.3934/DSFE.2024007
    [3] Juan Meng, Sisi Hu, Bin Mo . Dynamic tail dependence on China's carbon market and EU carbon market. Data Science in Finance and Economics, 2021, 1(4): 393-407. doi: 10.3934/DSFE.2021021
    [4] Xiaozheng Lin, Meiqing Wang, Choi-Hong Lai . A modification term for Black-Scholes model based on discrepancy calibrated with real market data. Data Science in Finance and Economics, 2021, 1(4): 313-326. doi: 10.3934/DSFE.2021017
    [5] Fatima Tfaily, Mohamad M. Fouad . Multi-level stacking of LSTM recurrent models for predicting stock-market indices. Data Science in Finance and Economics, 2022, 2(2): 147-162. doi: 10.3934/DSFE.2022007
    [6] Ming Li, Ying Li . Research on crude oil price forecasting based on computational intelligence. Data Science in Finance and Economics, 2023, 3(3): 251-266. doi: 10.3934/DSFE.2023015
    [7] Qian Shen, Yifan Zhang, Jiale Xiao, Xuhua Dong, Zifei Lin . Research of daily stock closing price prediction for new energy companies in China. Data Science in Finance and Economics, 2023, 3(1): 14-29. doi: 10.3934/DSFE.2023002
    [8] Georgios Alkis Tsiatsios, John Leventides, Evangelos Melas, Costas Poulios . A bounded rational agent-based model of consumer choice. Data Science in Finance and Economics, 2023, 3(3): 305-323. doi: 10.3934/DSFE.2023018
    [9] Tsumbedzo Mashamba, Modisane Seitshiro, Isaac Takaidza . A comprehensive high pure momentum equity timing framework using the Kalman filter and ARIMA forecasting. Data Science in Finance and Economics, 2024, 4(4): 548-569. doi: 10.3934/DSFE.2024023
    [10] Fangzhou Huang, Jiao Song, Nick J. Taylor . The impact of time-varying risk on stock returns: an experiment of cubic piecewise polynomial function model and the Fourier Flexible Form model. Data Science in Finance and Economics, 2021, 1(2): 141-164. doi: 10.3934/DSFE.2021008
  • In this paper, a linguistic steganalysis method based on two-level cascaded convolutional neural networks (CNNs) is proposed to improve the system's ability to detect stego texts, which are generated via synonym substitutions. The first-level network, sentence-level CNN, consists of one convolutional layer with multiple convolutional kernels in different window sizes, one pooling layer to deal with variable sentence lengths, and one fully connected layer with dropout as well as a softmax output, such that two final steganographic features are obtained for each sentence. The unmodified and modified sentences, along with their words, are represented in the form of pre-trained dense word embeddings, which serve as the input of the network. Sentence-level CNN provides the representation of a sentence, and can thus be utilized to predict whether a sentence is unmodified or has been modified by synonym substitutions. In the second level, a text-level CNN exploits the predicted representations of sentences obtained from the sentence-level CNN to determine whether the detected text is a stego text or cover text. Experimental results indicate that the proposed sentence-level CNN can effectively extract sentence features for sentence-level steganalysis tasks and reaches an average accuracy of 82.245%. Moreover, the proposed steganalysis method achieves greatly improved detection performance when distinguishing stego texts from cover texts.


    In the current context of global warming, the reduction of energy consumption is one of the central issues in research. Every day, part of the energy used is dedicated to the heating, ventilation and air conditioning (HVAC) of buildings. According to European Commission estimates, buildings in the EU are collectively responsible for 39% of our energy consumption and 35% of greenhouse gas emissions, which mainly stem from construction, usage, renovation and demolition [1].

    It is estimated that 75% of the building stock of 2049 is already constructed [2,3] and; yet, more than 35% of the housing stock is older than 49 years [4]. The main driver of energy efficiency in the dwelling sector is the renovation of the energy efficiency of the existing housing stock [5,6]. Additionally, renovation can also be a strategic driver of many transversal aspects.

    On the technical side, there are many methods of implementing technical solutions that lead to complex modifications in the interactions of the different parameters within buildings. Changing materials, type of glazing and heating or ventilation systems will alter the building's thermal, hydric, acoustic and lighting balance. [7,8,9] These modifications have a direct impact on energy consumption and can modify the quality of the ambiance [8]. From an economic point of view, building modernization often enables a reduction of the cost of operating and maintaining systems [9]. Additionally, it can also increase the financial value of the property by more than 20% [9]. To deal with the economic consequences of the recent COVID pandemic, the European Union proposed a resilient strategy to make cities more sustainable, create jobs and improve people's quality of life [10]. Beyond this economic aspect, there are additional societal issues regarding this wave of renovation and the housing affordability crisis [11].

    However, the total number of residential buildings renovated in Europe each year represents between 0.4% and 1.2% of the housing stock, for an average energy improvement of 9% [2]. The objective of this paper is twofold: first to provide a state-of-the-art analysis of the barriers to renovation, and, second to explore OSS models.

    In practice, introducing an additional actor to facilitate coordination among the parties involved, reevaluating the renovation process, or disseminating information also presents a new set of risks [12].

    Due to this, not all obstacles can be eliminated, and practices may differ to suit the target group. For instance, if administrative hurdles are prominent, the creation of facilitation-type OSSs may be necessary. On the other hand, for technical challenges, comprehensive OSS solutions that offer work supervision and monitoring services tend to be more effective [13].

    The objective of this work is to present the current context and highlight how to position oneself most effectively to accelerate the pace of renovation. By identifying and addressing the specific needs and obstacles within each context, the aim is to promote efficient and streamlined renovation processes.

    This work was carried out during the creation of the I-HEROS OSS. This is part of the plan to reduce CO2 emissions. In 2011, the General Council for the Environment and Sustainable Development established that on the French territory, greenhouse gas emissions must be divided by 4 until 2049 [13].

    Several documents in the scientific literature indicate that cost-effective energy measures are not always implemented. This gap between optimal implementation and actual implementation is called the energy efficiency gap or energy paradox [15,16]. The renovation process consists of a series of complex operations that involve many stakeholders or procurers who may be more or less integrated in the project path [5,18]. All of the mechanisms that reduce the efficiency of the results of a renovation, either technical, societal, economic or energetic, can be defined as obstacles to building refurbishment [18,20]. These different barriers are divided into five main categories and summarized in the Table 1.

    Table 1.  List and referencing of renovation barriers.
    Barrier Reference
    Organization of the market
    Fragmented market [18,20,21]
    Disjointed or lack of renovation support [18,20,22,23,24,25,26]
    Lack of time management / Planification [18,20]
    Availability of appropriate materials [19,24,25]
    Lack of or inadequate legislative framework / certifications [19,22,25,26]
    Lack of standardized solution [21,24,25]
    Company size [19]
    Skill and knowledge of professionals [13,14,15,16,17,18,19,20,21,22,23,24,21,26]
    Customer's lack of trust [13,21,24,25]
    Information
    Information quality [18,20,23,24,25,26,27]
    Information availability [19,22,23,24,25,26,27]
    Insufficient consideration of the impact of the occupant on the building [19,25,26]
    Insufficient integration of cross-cutting benefits [12,21,22,23,24]
    Sociological factors
    Lack of shared goals [18,20,21]
    Other priorities [18,20]
    Inertia to improve the sector [18,20,24,25]
    Decision making [18,20,22,23,24,25]
    Personal views [19,23,24,25]
    Technical
    Difference between expected and actual consumption [18,20,23,26]
    Project complexity [19,22,23]
    Long project duration [19,21]
    Administrative complexity [19,24,26]
    Complexity or lack of technical and managerial skills [19,22,26]
    Financial
    High-cost investments [18,20,22,23,24,25,26,27]
    Long payback period [18,20,23,24,25,26]
    Lake or difficulties of financing mechanism [21,22,23,24,25,26,27]
    External risks: energy price fluctuations [18,22,24,25]

     | Show Table
    DownLoad: CSV

    Market Organization: From a homeowner's perspective, the market for proposed services and renovation professionals is often perceived as fragmented and complex. The lack of support and understanding of the overall process can lead to demotivation and mistrust. Poor communication regarding project feedback can further exacerbate this mistrust. From the professionals' point of view, a lack of either qualified employees or materials can serve as barriers to the pace of renovations. The size of companies can also impact their long-term survival. The absence of standardized solutions necessitates a diversification of knowledge, potentially resulting in time loss during project execution. Moreover, the legal and regulatory framework significantly influences renovation incentive policies, market organization, fragmentation, and access to financial resources. The state and governmental authorities play a crucial role in expediting the renovation process.

    In a 2021 article, Pardalis et al. described the existing mechanisms for organizing the renovation market, highlighting the importance of OSSs as a key solution [20].

    Sociological factors: From a sociological point of view, they play a significant role in the decision-making process, making it challenging to provide accurate information that can facilitate this process. This challenge is further exacerbated in the case of co-ownerships, where not all occupants necessarily share the same objectives, expectations, priorities, or understanding of the renovation project. Additionally, personal affinities can complicate the situation.

    Technical: From the owner's perspective, the fragmentation of the market and the legal framework can result in a complex administrative process. Following this stage, the management and organization of the renovation work require specific skills in project management. Additionally, there can be discrepancies between the expected and actual results. These factors are particularly crucial in complex renovation projects, such as historical or classified buildings. Often, owners must adapt a building that was originally designed to meet prior needs to address current issues, techniques, and materials. Failure to adequately consider this aspect can lead to significant disruptions on the construction site, building, and maintenance, including common issues such as ventilation problems, infiltration, or structural weaknesses [8].

    Information: The quality and availability of information is a determining factor in the motivation and success of a renovation project. Just like the quality of the communication to the occupants, it allows for the understanding of the technical, financial, organizational. points necessary to the decision making and to the good progress of the renovation process. For example, a poor understanding of the occupants and their habits can lead to the choice of an inappropriate renovation program.

    A good communication on the transversal energy improvements can also motivate the decision process and accelerate the renovation process. Therefore, it is important to highlight the benefits in terms of asset enhancement, comfort improvement, financial gains on operations and maintenance.

    Financial: The quality and availability of information are crucial factors in motivating and ensuring the success of a renovation project. Similar to effectively communicating with the occupants, it enables a comprehensive understanding of the technical, financial, and organizational aspects necessary for decision-making and the smooth progress of the renovation process. For instance, a lack of understanding about the occupants and their habits can result in the selection of an inappropriate renovation program.

    Furthermore, effective communication regarding cross-cutting energy improvements can also enhance the decision-making process and expedite renovations. It is essential to emphasize the benefits in terms of asset enhancement, improved comfort, and financial gains in operation and maintenance.

    By providing comprehensive and clear information, stakeholders can make informed decisions and understand the positive impact that renovations can have on various aspects, thereby fostering motivation and facilitating the renovation process.

    An OSS can be described as a virtual and/or physical hub where owners can access all the information and services, they need to undertake their project [27]. In the context of an energy renovation, it provides a comprehensive package of services to assist them in managing all the necessary steps of a renovation project. Irrespective of the specific administrative structure in place, this single access point serves as the interface between the owners and the service providers. [25,29,30,31].

    Various initiatives can effectively address the obstacles of renovation. Some initiatives can be seamlessly integrated into the process, such as Energy Service Companies (ESCOs) or what Mandy de Wilde and Spaargaren. refer to as "strategic intermediaries design services" [30]. Additionally, there are initiatives that are less intertwined with the renovation process, such as specific measures implemented by public authorities. For instance, Germany's Kreditanstalt für Wiederaufbau (KfW) programs address financial concerns, while the Sustainable Energy Action Plan (SEAP) offers comprehensive design solutions.

    Figure 1.  Difference of organization of the renovation market, based on [29] and [33].

    This article specifically focuses on the OSS case. When it comes to automating renovation support and expediting the pace of renovation, OSSs prove to be significantly more effective. They are the primary driver for governments when implementing such services [31]. Additionally, even if these kinds of processes can solve many barriers to the renovation, the use of such complex organizational systems can generate the development of new barriers [12]

    Brown et al. proposed a representative example of what an OSS is and how it supports the customer for avoiding risks and failures related to the current market [29,34].

    OSS reduce the frequency of interactions between different stakeholders, as well as improve access to services. The level of integration and execution of the renovation process can differ from one structure to another. The key to establishing a successful OSS is to identify the appropriate service packages that align with the specific needs of the territory. Consequently, some initiatives may be more closely aligned with the regulatory framework, while others may be tailored to provide commercial considerations.

    In the literature, four different types of one stop shop business models have been identified. They proposed different levels of support and operating systems [25,29,30,31,34].

    The main objective of this OSS is to provide assistance to homeowners and promote renovation. In this model, advisors offer general guidance on the scope of the work or project direction. First, the services focus on raising awareness about the benefits of renovation. As part of this, the OSS may or may not provide a list of local suppliers and service providers. This model is particularly suitable for homeowners who are motivated and capable of organizing and overseeing the project but are seeking a reliable source to find the missing information or guidance.

    In a 2021 article, Bagaini et al. [21] proposed a subclassification of the first three types of OSS. In the first category of facilitation, the OSS of the (i) advice type has basic services oriented on counseling. They can either be general or personalized advising according to the occupants' personal or public data. Online platforms are highly suitable for this kind of initiative. (ii) An intermediary OSS with the same objectives can offer homeowners more support regarding financial or technical aspects. This business model works best when aligned with governmental structures, associations or facilitators. The last sub-category of OSS, the (iii) Lender type, specifically offers financial services with the option of organizing a schedule for payments. This aspect can either be handled directly by the companies themselves or by involving bank partners within the OSS. The involvement of financial institutions in providing support can have implications for the responsibility and quality of the services offered, whether they are subcontracted or not. These initiatives can be driven by governments, cooperative structures, or facilitators, depending on the specific context and setup.

    In this second model, the OSS is established as an additional component of an existing structure. This service takes a comprehensive approach that includes organizing and facilitating access to all the services available within the territory. The OSS goes beyond providing simple advice and can assume the role of project supervision. However, it should be noted that the responsibility for the successful completion of the work ultimately lies with the companies involved, rather than the one-stop shop.

    In the second type of model, the OSS coordination model, Bagaini et al. [21] proposed a segmentation into two categories: connection and cooperation services. In the first case, the (iv) connection model positions itself in the market as a trusted team or network capable of addressing the various needs and aspects of the renovation process. This type of OSS is commonly developed by associations or groups of building professionals, although it can also be led by governments or facilitators. The second type of OSS, the (v) cooperation type, can offer a work realization service. Most often, a team of multi-skilled professionals will join together to provide complete packages of works.

    In this particular model, the OSS assumes full responsibility for the entire renovation process, starting from the project definition and continuing until the delivery of the completed building. It ensures the full continuity of the renovation journey for the owners. In the event of any differences between the initially planned outcomes and the actual results, the OSS will hold the owner responsible for resolving any issues and addressing any concerns.

    For the third kind of OSS, all-inclusive models, Bagaini et al. [21] described two types of subcategories. In the (vii) project management type of OSS, the OSS often acts as a project manager or construction manager's assistant. It subcontracts the work, though it is responsible for the suppliers, companies, and assists in the preparation of the subsidy and financing files. It is also responsible for the deadlines, the guarantees and the quality of the work. This type of OSS is most often carried by professional cooperatives, municipalities or facilitators. The last sub-category of OSS corresponds to the (viii) full implementation model. The operation is very similar to the project management model except that all activities, skills and resources are internalized in the OSS. These are "turnkey" renovation solutions.

    In this model, the one-stop shop assumes responsibility for ensuring the smooth progression of the customers journey and the successful completion of the renovation work. Additionally, the OSS is dedicated to achieving specific energy savings targets. Pardalis et al. define this model as a hybrid approach that combines the energy performance assurance typically offered by Energy Service ESCOs with the energy efficiency measures provided by the OSS. [25]. The support to the renovation is mainly performed through specific energy contracts: for example, energy performance contracts or energy supply contracts.

    Similar to the all-inclusive OSS models, also provide comprehensive solutions. However, they involve the inclusion of a third party, known as the ESCO, acting as an intermediary between the individual homeowner and the OSS. Through an agreement, the ESCO outlines different measures that the homeowner must adhere to in order to enhance the quality of the renovation process [12]. This approach is particularly well-suited for OSS that handle a large volume of similar projects, such as retail stores, industrial facilities, or municipalities. Additionally, there are various assembly options available within this framework. Brown et al. [33] and Hannon et al [33] detailed some of them in their respective articles.

    The organizations or services provided by OSS can also be categorized based on the entities responsible for overseeing the project. The management structure, funding sources, interests, and scope may vary, resulting in different approaches to providing support. As a result, there is a diversity of services offered within different types of OSS models, reflecting the varying approaches and capabilities of the providers [13,35,36].

    Government-driven or locally/regional municipality-led: These OSS receive support from either government entities or local authorities. As a result, private actors are assisted by municipalities in carrying out the renovation projects. The financing for these initiatives can be entirely public, sourced from either citizens' taxes or European funds, or partially public. In some cases, certain services are provided by private companies in exchange for compensation, such as project setup and work management. Generally, the motivation behind establishing and managing these OSS is driven by climate-related, social, or energy resource management objectives. Figure 2 illustrates the relationship between the stakeholders for the municipal OSS.

    Figure 2.  Schematization of the process of a governmental driven one stop shop.

    Industry driven: An OSS operated by an industry player can be exemplified by a building industry company that has developed a specific technical renovation solution. The focus of this type of OSS is to promote and advocate for their own market solution, aiming to expand their market presence and increase the adoption of their renovation solutions. Since local municipalities often strive to enhance the quality and accessibility of renovation solutions within their territories, they may support and incentivize these initiatives through subsidies or other means. [25,36]. The operating funds of this type of OSS can be diverse. Figure 3 illustrates the relationship between the stakeholders for the industry driven OSS.

    Figure 3.  Schematization of the process of an industry driven one stop shop.

    ESCO based: Hannon describes ESCOs as companies that provide energy to homeowners [33]. As mentioned earlier, support for renovation projects is primarily facilitated through specialized energy contracts, such as energy performance contracts or energy supply contracts. These models are often intricate and necessitate the involvement of technical partners to rationalize the renovation process. Figure 4 illustrates the relationship between the stakeholders for ESCO OSSs.

    Figure 4.  Schematization of the process of an ESCO one stop shop (Brown, 2018).

    Independent facilitators: In OSS operated by facilitators, the level of support or assistance provided can vary significantly. Facilitators act as intermediaries between contractors and homeowners, aiming to streamline the interaction and facilitate the renovation process. Facilitators assist with financial arrangements for projects, participate in specific project tasks, and offer technical or legal advice. Their primary focus is to provide support services for residential renovations. In cases where facilitators are entrepreneurs or consultants, their objective may be to either grow their existing business or expand their service offerings to cater to their clients' needs. Figure 5 illustrates the relationship between the stakeholders for the facilitation OSS.

    Figure 5.  Schematization of the process of facilitators one stop shop [25,34].

    Cooperative/multi-disciplinary team: The OSS is established by a team of professionals collaborating to enhance their collective expertise. They position themselves in the market as a unified entity providing a range of modular, plug-and-play services tailored to meet homeowners' needs. These teams consist of various building industry professionals, including architects, manufacturers, energy auditors, and more.

    According to Bertoldi et al. [13], the main objectives are not necessarily focused on energy or financial savings. Additionally, they can prioritize cross-cutting objectives. The range of services offered extends from basic renovation operations to comprehensive and intricate work programs. Figure 6 illustrates the relationship between the stakeholders for the cooperatives OSS.

    Figure 6.  Schematization of the process of cooperatives/ multi-disciplinary team one stop shop [25,33].

    Store: This type of OSS is primarily intended for large retail stores or warehouses. Its main objective is to offer and promote solutions, technologies, and building materials. The OSS serves as a point of sale where customers can receive specialized technical advice on practical aspects of construction works. This type of OSS can also correspond to those that Pardalis and Georgios [25] described as complementary business. It attracts cross-sector actors who aim to tap into the renovation market and utilize the OSS to structure their offerings. Examples include insurance companies, utilities, and more.

    Digital tool: Pardalis and Georgios described this particular business model as an automated support process. This type of OSS model relies on gathering information from the homeowner and subsequently offers various tailored renovation programs aligned with the project's requirements. [25]. It also serves as a central hub where homeowners can access all the information required for the smooth execution of their renovation works. Figure 7 illustrates the relationship between the stakeholders for the digital OSS.

    Figure 7.  Schematization of the process of digital one stop shop (Pardalis, 2021).

    Finally, the size and scope of operation also have an impact on the services offered and the type of OSS. Depending on the scale and the number of homeowners within a territory, the support strategies, types of one-stop shops, and funding mechanisms may vary. Additionally, it is possible to have multiple one-stop shops operating within the same territory. In such cases, the services offered by these different entities can be complementary. Taking into account these three factors that can influence the services provided by an OSS, Figure 8 summarizes the organization of the renovation market across the European territory.

    Figure 8.  Summary of the organization of the renovation market on the European territory.

    The different levels of territories that a OSS can cover are as follows [13,15,29]:

    European level: At the European level, there are opportunities for financial support and research initiatives related to an OSS and renovation policies. The objective of these selected projects is to contribute to the reduction of greenhouse gas (GHG) emissions and achieve climate objectives.

    National level: National solutions come in various forms, with the most common being services linked to national subsidies to incentivize individuals to renovate. For instance, energy saving certificates are a program that benefits households. Since 2016, certain companies have been able to purchase a portion of energy savings achieved through renovation work from households experiencing energy insecurity [37]. Often on a national scale, due to the large number of households and the size of the territory, OSS often provide either advisory services or financial incentives for renovation projects.

    Regional scale: At the regional level, the political framework plays a significant role in determining the pace and approach to renovation. OSS at this scale are better positioned to offer more technical services and to oversee the progress of renovation works. This level often offers a broader range of services, including advice, project supervision, and organization of the professional and material supply chains within the region.

    Local scale: OSSs managed by municipalities are prevalent at the local level, as municipalities are key stakeholders within their respective territories. Since this level has the smallest geographical segmentation, it is important to provide complementary services that align with existing offerings in the area.

    OSSs and integrated renovation support services play a crucial role in accelerating the pace of renovation. As described earlier, the various business models of one-stop shops, as described earlier, contribute to reducing barriers in different ways. However, the primary objectives of all OSSs are to serve as intermediaries between homeowners and professionals and/or to organize the renovation market.

    By doing this, OSSs help improve the overall organization of renovation projects, streamline the supply of materials, and ensure the proper execution of work techniques, among other benefits. Additionally, the impartiality of the advice provided also enhances the decision-making process, leading to better outcomes.

    The facilitation model effectively addresses certain barriers in the renovation process. By focusing on the orientation stage, this model particularly helps both information and social barriers. Having a single point of contact and easy access to information significantly reduces information barriers.

    For instance, a notable example of the facilitation model is the national-scale project "France Rénov" led by the National Housing Agency in France. Another approach is the purchase of energy savings by distribution companies. In addition to addressing information barriers, the facilitation model aims to raise awareness about the benefits of energy savings and reduce social barriers. However, a lot of other barriers remain, which can lead to the decision to not implement the energy renovation process [29].

    The coordination model effectively eliminates multiple barriers in the renovation process. By providing comprehensive support throughout all stages of the renovation, the OSS helps overcome political/institutional, social, and market barriers. Through the establishment of a reliable network, the coordination model bridges the gap between the fragmented supply chain and the demand side. Acting as a single point of contact, the OSS significantly reduces market barriers. This an integrated approach enhances coordination, streamlines communication, and fosters trust, resulting in a positive impact on the renovation process [29].

    The coordination model significantly contributes to achieving good quality renovations and realizing the estimated energy savings. It effectively reduces the stress, disruption, and issues typically associated with renovation projects [28]. This has impact on addressing social barriers.

    The I-HEROS OSS in Toulouse fits perfectly into this category, which aligns perfectly with this approach. Operating at multiple levels, it leverages existing support channels within the territory to guide occupants based on their specific needs.

    Additionally, the coordination model offers advantages for the supply side of the renovation industry. First, the OSS serves as a lobbying group for suppliers, facilitating connections between suppliers and customers, as well as fostering collaborations and knowledge transfer among suppliers themselves. This coordinated collaboration encourages the exchange of expertise, skills, and innovations, leading to positive outcomes [29].

    The development model has a notable impact on various barriers, including economic, social, technical, and market barriers. By ensuring the achievement of high energy savings and facilitating access to financial support, it effectively reduces social and economic barriers. Homeowners become more aware of the benefits of renovation, making the decision-making process easier and reducing the stress, disruption, and hassle typically associated with renovation projects [29].

    One of the key advantages of the development model is its ability to mitigate market fragmentation and push trust among stakeholders. With limited influence from multiple stakeholders, the model promotes a more cohesive and coordinated approach, reducing information discrepancies, and increasing trust; this positively affects market barriers.

    This part focuses on different case studies of OSS in Europe and the type of business model associated with them.

    A benchmark of existing OSSs in Europe identified 63 case studies has been produced. This exercise illustrates the functioning and the articulation of the different OSSs between them. There can be different OSSs in the same territory that respond to different problems and complete each other. This is most apparent in the western European regions where the number of identified cases is highest.

    Figure 9.  Location of identified OSS across the Europe.

    All 63 OSSs studied come from research projects [38], with documents provided by the European Union [12,39], the ADEME (French Association for the Environment and Energy Management) [39] and by the results of the study during the setting up of the I-HEROS OSS.

    By cross-referencing the information in the literature, we have proposed a classification of one-stop shops into four types of business models corresponding to seven types of support services. Finally, the last level of segmentation of the renovation market identified is the field of application: local, regional, national, or European. The benchmark study revealed that different OSS can be present in a single territory.

    Therefore, the services offered by the different types of business models are complementary. Generally, the greater the scope of the OSSs, the less operational the service is.

    The benchmark was based on comparison criteria and tables similar to those used by the framework of Cicmanova and Maraquin. [39] and Bertoldi et al. [13]. Based on the services provided, the study of the customer path, leads to the selection of an OSS class.

    One of the conclusions of these same authors is the lack of maturity and resources/ambition for coordination and facilitation models. However, the graph in Figure 10 shows that the different types of OSSs are represented in almost equal numbers across the territories. This clearly demonstrates the existing need and the interconnectivity of the need for support in renovation.

    Figure 10.  Distribution of the types and business models of one stop shop in Europe.

    The majority of identified OSS are located in Eastern and Western Europe, covering a wide range of territories. All types of OSS models are represented in these regions. The most prevalent models found are connection, advice, and project management models. In Eastern and Southern Europe, the majority of one-stop shops operate under facilitation models, often functioning as either lenders or intermediaries. In Northern Europe, the majority of OSS reported are of the coordination or facilitation type, with either intermediate or cooperative structures.

    It is also interesting to observe the nature of the OSS according to their type. The more facilitation-type activities they have, the larger is their geographical coverage. They are also frequently financed by government entities. Inversely, the more support and follow-up services they provide, the more private they will be, and the more limited their number of renovation projects. The most advantaged countries are those with a flexible governmental context, allowing for both extremes [28].

    Several factors may contribute to the variation in market organization in Europe, such as the level of building consumption and density in a particular territory [42], as well as the type of governance structure in place [43]. While the majority of identified OSSs are located in Western Europe, variations can be observed in their operations within these regions. When comparing the annual renovation rates to European data, the rates remain relatively similar in magnitude [2].

    This points up the disparity of support needs in the different European territories and the impact of cross elements. The selection of a OSS model cannot be solely based on factors like scope, project ownership, or the presence of other existing one-stop shops. It should also take into consideration the opportunities and specific needs within the region into consideration.

    Zimmermann and Feiertag's book [43] explained the importance of territorial governance issues and how they can affect the development of an urban planning strategy. The article by Foster and Saleh illustrated these disparities through their study of urban planning policies on the circular use of buildings in different European countries [44]. For this specific example, western and northern Europe are way more dynamic.

    On another matter, Betti and Gianni, through several indexes of living standards, described that there are important variations of poverty on the European ground that can exceed 15% of variation [45]. This can be also an element of explanation on the difference of management linked to the accompaniment of the renovation.

    For instance, the findings of the European LOCATE project have revealed variations in the overall energy consumption mix, residential consumption patterns, and access to renewable energies across different countries [42]. Additionally, differences in climate conditions have resulted in varying expectations for comfort and corresponding architectural strategies [46].

    For example, Manzano et al. propose a review of bioclimatic architectural strategies and issues. This article highlighting the significant influence of climate on construction methods. Furthermore, the characteristics of residential building stocks can also differ among countries. For example, Marinova et al. provided insights into the ratio of collective housing to individual housing by region and the materiality based on the age of the building stock [47].

    These factors are unique to the historical context of each territory, and as a result, the renovation solutions will not be necessarily identical. These differences inevitably give rise to diverse needs in terms of renovation works and guidance. Considering the diverse and heterogeneous nature of these territories, the strategies for establishing OSSs must be adaptable.

    Taking this analysis into account, one approach to assess the replicability of a OSSs could involve examining the similarity of dynamic criteria associated with renovation within a specific territory. These criteria could include quality of life, energy accessibility, territorial governance, climate conditions, energy consumption, patterns, urbanization levels, and more. By evaluating these factors, it becomes possible to gauge the feasibility of replicating a successful OSS model in a different location.

    According to the Pardalis and Georgios Swedish study, enhancing the energy-saving renovation market requires considering the current market conditions while also developing strategies to facilitate its evolution. Given that the OSS market already exists and is well-established, one approach to expediting the renovation process would be to implement targeted initiatives. This could involve identifying critical residential properties based on specific criteria and offering specialized support programs tailored to their needs. By focusing efforts on these key dwellings, the pace of renovation can be significantly accelerated [25].

    As previously discussed, the low renovation rates can be attributed to various barriers, and OSS offer a potential solution to overcome these obstacles. Additionally, we have demonstrated that the adoption of a renovation strategy should be linked to the specific needs of each territory, which accounts for the divergent support strategies employed by an OSS.

    By analyzing data from the European Union, we compared the renovation rates of different countries with various territorial indicators (Table 2 and Figure 11). Through this analysis, we identified the factors that carry the most significant weight and exhibit strong correlations. Focusing on these key factors within the OSS would result in an enhanced efficiency and effectiveness in addressing the renovation challenges. This work also sheds light on the factors that could accelerate the pace of renovations in their respective territories.

    Table 2.  Description of the indicators selected for the correlation matrix.
    Data type Various indicators Source
    Energy renovation in residential buildings Below threshold energy renovation: Savings gain < 3% Light energy renovation: Savings gain 3% ≤ x ≤ 29% Medium energy renovation: Savings gain 29% < x ≤ 60% Deep energy renovation: Savings gain > 60% [2]
    Repartition of the years of residential construction Share of construction before 1945 (%) Share of construction betwen 1945–1980 (%) Share of construction betwen 1981–2000 (%) Share of construction after 2001 (%) [48]
    Share of final energy consumption in the residential sector by type of end-use Share of final energy consumption (%): space heating Share of final energy consumption (%): space cooling Share of final energy consumption (%): water heating Share of final energy consumption (%): cooking Share of final energy consumption (%): lighting and supplies Share of final energy consumption (%): other end use [49]
    Share of fuels in the final energy consumption in the residential sector Share of solid fossil fuels in final energy consumption (%) Share of natural gas in final energy consumption (%) Share of oil and petroleum products in final energy consumption (%) Share of renewables in final energy consumption (%) Share of electricity in final energy consumption (%) Share of heat in final energy consumption (%) [49]
    Electricity price Electricity price (€/kWh) [50]
    Climate Average heating degree days 1979–2022 Average cooling degree days 1979–2022 [51]
    Population structure, by urban–rural typology Predominantly urban regions (%) Predominantly intermediate regions (%) Predominantly rural regions (%) [52]

     | Show Table
    DownLoad: CSV
    Figure 11.  Renovation rate by European country.

    When comparing the renovation rates of various countries and the presence of OSSs in their territories, it becomes evident that advisory models prove effective for minor actions, while a network of interconnected OSS at different levels can yield more substantial results in terms of deep renovations. Figure 11 also indicates that the mere presence of a OSS cannot entirely account for its effectiveness. In fact, there are instances where only a few OSSs have been identified, yet the renovation rate surpasses that of other countries (e.g., Bulgaria, Finland, Croatia, etc.).

    Additionally, Figure 11 also reveals an imbalance between the number of identified OSSs and the corresponding renovation rates for each country. Renovation rates aimed at achieving energy savings between 3% and 30% do not proportionally correlate with the number of OSSs identified, whereas for operations exceeding 30%, they show greater significance. This suggests that the OSS models identified might not be as effective for renovations falling between 3% and 30%.

    Moreover, variations in the types of models identified for different countries indicate that regions adopt diverse approaches for renovations exceeding 30%.

    Correlation refers to the relationship or dependence between variables. When the distribution of data points of one variable in relation to another follows a linear pattern, the variables are considered to have a linear correlation. The strength and direction of this relationship can be quantified using the Pearson correlation coefficient. This coefficient is derived from the covariances of the variables divided by the product of their standard deviations. A coefficient close to either 1 or –1 indicates a strong positive or negative correlation, respectively, implying a proportional or inverse relationship between the variables. The Figure 12 details the correlation between the indicators selected in Table 2.

    Figure 12.  Correlation matrix between renovation rate and territorial characteristics.

    For buildings constructed before 1980, there is a stronger correlation between the consumption of heating (0, 6), hot water (0, 33), and lighting (0, 16). Additionally, heating consumption represents a higher proportion compared to lighting or hot water consumption across all categories. From an energy standpoint, the main correlation with heating usage is gas (0, 53). These buildings from the pre-1980 era hold significant potential for renovation.

    A notable correlation between different energy consumption categories can indicate the potential for comprehensive renovation projects. Energy consumption is also closely tied to climatic conditions (more than 0, 5). There is a correlation observed between electricity usage in urban areas and the demand for air conditioning. Urban areas show a higher share of electricity consumption (0, 18), particularly in newer buildings (0, 32). Additionally, the use of renewable energy sources is more prevalent in rural areas (0, 52).

    In the context of renovation, the primary correlations are observed between lighting and heating consumption. While there is a decrease in the correlation between heating consumption and energy efficiency improvements, there is an increase in the correlation with air conditioning consumption. As a result, it is crucial to pay attention to renovation methods to prevent a deterioration of thermal comfort and the reliance on air conditioning.

    The first section of this article provides a comprehensive summary of the existing literature on the topic of renovation disincentives. Additionally, it includes diagrams of the different OSS options available in the renovation market. This work serves as a state-of-the-art review of current practices, focusing exclusively on OSS solutions. While certain authors do reference alternative approaches that focus on specific aspects of the renovation process, OSS stand out as the most effective remedies [20,36]. Therefore, to further advance this work, integration of systems beyond OSSs could be explored. While an OSS proves highly effective in the context of overall renovation efforts, there is potential for incorporating specific targeted operations, as discussed in the preceding section. For example, incorporating disparate segments into an OSS, rapid and efficient outcomes can be ensured.

    Furthermore, the establishment of new OSSs can give rise to the emergence of additional challenges in the field of renovation. Therefore, conducting a comprehensive examination of these new obstacles would be of great interest. For instance, how can we ensure that the digitalization of these services is accessible to all [53,54]? Additionally, how can we explore alternative business models to reduce reliance on public funding, without compromising the integrity or quality of the advice provided?

    The objective of the remaining sections is to compare the theoretical assumptions identified in the literature with the actual operations of OSS, focusing on the existing initiatives in Europe. One limitation of this study is the reliance on referencing projects associated with national or European funding programs. Consequently, smaller-scale initiatives, towns without access to such programs, or OSS models not subject to European directives are not considered in this benchmark. While our research demonstrates the effectiveness of these OSS for large-scale renovations, questions remain about their support for operations that do not achieve energy savings of < 30%. Figure 11 illustrates this concern. Although the number of OSSs surveyed does not adequately represent the quantity of renovations between 3% and 30%, it is more representative for renovations below 30%. Another avenue for improving this work could involve conducting a more in-depth analysis of the identified OSS benchmarks. Creating a comprehensive list of active and methodology OSSs and examining their respective challenges and successes would be instrumental in refining existing solution.

    The final point of discussion revolves around the quality of the data utilized to characterize the territories and estimate variations in the support requirements across different cities. This analysis relies on indicators provided by the European Union, making it intriguing to compare these outcomes with other demographic and geographic data to validate the results.

    This article details the current state of the structure of the renovation market in Europe, its problems, its solutions, and its challenges for understanding or building up a support service for housing stock. While the renovation of the housing stock seems to be an effective solution to reach the 2049 greenhouse gas emission targets, the rate and quality of the work done until now is still too low.

    First, this article proposes a state-of-the-art and a brief description of the barriers to renovation. The literature review identifies 26 major issues, classified into five categories: sector organization, information, sociological, technical, and financial. These obstacles are mainly due to human and logistical factors in the renovation process.

    The following section highlights OSS as a viable solution to these challenges, providing a virtual and/or physical hub where owners can access comprehensive information and services required for their renovation projects. While there are alternative approaches to integrating a renovation support system, OSS emerges as the most effective solution.

    Drawing on previous literature, this article proposes a classification framework, to depict the distribution of these OSSs across different territories. The classification takes into account their scope of operation (local, regional, national, or European), their ownership (governmental authorities, associations, cooperative industries, stores, individual or multiple facilitators), and the types of services they offer (advisory, intermediary, lender, connection, cooperation, project management, all-inclusive services).

    A comprehensive benchmark of 63 OSSs in Europe was conducted, allowing for a comparative analysis of the distribution of OSS models across various countries and territories. This examination revealed that selecting an appropriate OSS model cannot be solely based on factors like scope, ownership, or the existence of other OSSs. It necessitates a consideration of the unique opportunities and needs present within each region.

    This study has also emphasized the necessity to diversify our support methods, as the majority of existing literature concentrates solely on government-driven or nationally directed initiatives. Additionally, we have demonstrated in this article that while such initiatives are effective in ambitious renovations, they might not accurately represent the majority of renovations undertaken thus far

    Factors such as urban planning conditions, prevailing governance structures, climate characteristics, and the existing built heritage play a significant role in determining the most suitable OSS model for a particular region. Each locality presents its own set of distinct circumstances that must be considered when designing and implementing an effective renovation support system.

    Furthermore, we have observed a significant number of one stop shops and support services in countries with cold climates, where there exists a high potential for energy savings. It is equally important to explore regions with lower heating consumption levels.

    In the final section, a comparison is made between several criteria that have been identified as potentially indicative of renovation activities. These criteria include energy consumption and type, urbanization rate, construction date, and climatic context. These factors are then compared with the renovation rate to establish any correlations or patterns. This study report contributes to the broader understanding of building energy renovation activities and the uptake of nearly zero-energy buildings in the EU [2].

    The authors declare that they did not utilize Artificial Intelligence (AI) tools in the creation of this article.

    All authors declare no conflicts of interest in this paper.

    This work was financed by the I-Heros project that has received funding from the European Union's Horizon 2020 research and innovation program under grant agreement No 890598 and by the ADEME [41].

    Table A1.  Benchmark of European one-stop shops and their management.
    Facilitation model Coordination model Developement model
    Advisory model Intermediary model Lender model Connection model Cooperation model Project management model Full implementation model
    France
    SEM ile−de−france Energie 1 Ile−de−France 1
    Renovation nouvelle aquitaine 2 Nouvelle Aquitaine Region (Bordeaux) 1
    Picardie pass renovation 3 Picardie région (individual houses)/
    Hauts−De−France (Condomniums)
    1
    Oktave 4 Alsace Champagne, Ardenne Lorraine
    region (strasbourg)
    1
    Bordeaux Métropole Energies (Facirénov) 5 Bordeaux 1
    Région Normandie OSS 6 Normandie 1
    Brest Métropole OSS (Tinergie) 7 Brest Region 1
    Espace−info Energie 8 Toulouse Metropole 1
    Energiesprong 9 different regions in
    France, Ile−de−France
    1
    AREC Occitanie 10 Occitanie Region 1
    Belgium
    Homegrade 11 Brussels Capitol
    region
    1
    Huisdokter 12 Limburg 1
    Renowatt 13 Liege 1
    Vlaams energiebedrijf 14 Flanders, Antwerp 1
    CLEAR 15 Brussel 1 1
    Germany
    Energiesprong managed by DENA 16 Berlin 1
    Bauteam Hamburg 17 Hamburg city 1
    Denmark
    Frederikshavn OSS 18 Frederikshavn Municipalty 1
    CLEAN Green Business Growth 19 4 cities, Region Syddanmark 1
    BedreBolig 20 Denmark 1
    BetterHome 21 Denmark 1
    CleanTech 22 Denmark
    ProjektLavenergi By ADSBOLL 23 South Denmark, mainly Kolding 1
    Sustain Solutions 24 Denmark 1
    Zerohome Program 25 Sonderborg 1
    Finland
    ENRA concept 26 Helsinki 1
    Norway
    Bolig Enok 27 Oslo 1
    Estonia
    TÜ Energiaühistu 28 Tallinn 1
    Tartu Regional Energy Agency 29 Tartu 1
    Vaxjo OSS 30 Kronoberg 1
    KredEx 31 Tallinn 1
    Hungary
    RenoHUb 32 Budapest 1
    Projekt Doktor 33 Budapest 1
    Netherlands
    Center for sustainability 34 Amsterdam 1
    Reimarkt 35 Bergen op zoom 1
    WoonWijzer−Winkel Rotterdam 36 Rotterdam, Limburg 1
    Haarlemse Huizenaanpak 37 Haarlem 1
    Stroomversnelling 38 Several locations, Amsterdam 1
    Slim Wonen in Leeuwarden 39 Leeuwarden 1
    Slovakia
    EBRD credit lines 40 Various places in Bulgaria (also Slovakia), Sofia 1
    FinEERGo−Dom 41 Sofia, but also in other countries like Poland, Latvia, Romania, Slovakia and Austria 1
    EEE Consortium 42 Sofia 1
    Rhodoshop Programme 43 6 municipalties from the Rhodope region. 1
    Cyprus
    Aradippou Municipality One Stop Shop 44 Municipality of Aradippou, Larnaca district 1
    Czech Republic
    SMART CITY 45 Litoměřice
    Litoměřice 46 Litoměřice city 1
    Italia
    Infoenergia 47 Province of Trento 1
    Project Development Assistance for
    Local Authorities in Piemonte
    48 Piemonte region 1
    Mantova OSS 49 Mantova city 1
    Superhomes 50 Tipperary city 1
    England
    Retrofit Works 51 London 1
    Ecofurb(collaboration with Retrofit works) 52 London 1
    Allenergy's Affordable Warmth Service 53 Argyll & Bute region, Scotland 1
    Tighean Innse Gall 54 Western Isles 1
    RenovationUnderwriting 55 London 1
    People Powered Retrofit 56 Manchester 1
    Greece
    Energy Communities 57 Attica 1
    Spain
    Officina de l'energia 58 Valencia 1
    Energy Housing Assessment Tool (ENERHAT) 59
    60
    Spain
    Spain
    1
    Ireland
    Energy Communities Tipperary Cooperative 61 Nenagh 1
    Croatia
    KLIK Energy Cooperative 62 Križevci 1
    Bulgaria
    IZGREI BG 63 Plovdiv 1

     | Show Table
    DownLoad: CSV
    Figure A1.  Territorial indicators by European country.


    [1] Y. L. Liu, H. Peng and J. Wang, Verifiable diversity ranking search over encrypted outsourced data, CMC-Comput. Mater. Con., 55 (2018), 37-57.
    [2] L. Y. Xiang, Y. Li, W. Hao, et al., Reversible natural language watermarking using synonym substitution and arithmetic coding, CMC-Comput. Mater. Con., 55 (2018), 541-559.
    [3] H. M. Meral, B. Sankur, A. S. Ozsoy, et al., Natural language watermarking via morphosyntactic alterations, Comput. Speech Lang., 23 (2009), 107-125.
    [4] C. M. Taskiran, M. Topkara and E. J. Delp, Attacks on lexical natural language steganography systems, Proceed. SPIE, 6072 (2006), 607209-607209-9.
    [5] Z. L. Chen, L. S. Huang, H. B. Miao, et al., Steganalysis against substitution-based linguistic steganography based on context clusters, Comput. Electr. Eng., 37 (2011), 1071-1081.
    [6] Z. L. Chen, L. S. Huang and W. Yang, Detection of substitution-based linguistic steganography by relative frequency analysis, Digit. Invest., 8 (2011), 68-77.
    [7] L. Y. Xiang, X. M. Sun, G. Luo, et al., Linguistic steganalysis using the features derived from synonym frequency, Multimed. Tools Appl., 71 (2014), 1893-1911.
    [8] L. Y. Xiang, J. M. Yu, C. F. Yang, et al., A word-embedding-based steganalysis method for linguistic steganography via synonym-substitution, IEEE Access, 6 (2018), 64131-64141.
    [9] Z. S. Yu, L. S. Huang, Z. L. Chen, et al., Steganalysis of synonym-substitution based natural language watermarking, Int. J. Mult. Ubiquit. Eng., 4 (2012), 21-34.
    [10] Z. S. Yu, L. S. Huang, Z. L. Chen, et al., Detection of synonym-substitution modified articles using context information, Second International Conference on Future Generation Communication and Networking, (2008), 134-139.
    [11] Y. T. Chen, J. Xiong, W. H. Xu, et al., A novel online incremental and decremental learning algorithm based on variable support vector machine, Cluster Comput., Available from: https://doi.org/10.1007/s10586-018-1772-4.
    [12] L. Y. Xiang, G. H. Zhao, Q. Li, et al., TUMK-ELM: A fast unsupervised heterogeneous data learning approach, IEEE Access, 6 (2018), 35305-35315.
    [13] I. A. Bolshakov, A method of linguistic steganography based on collocationally-verified synonymy, International Workshop on Information Hiding, (2004), 180-191.
    [14] C. Y. Chang and S. Clark, Practical linguistic steganography using contextual synonym substitution and a novel vertex coding method, Comput. Linguist., 40 (2014), 403-448.
    [15] X. Yang, F. Li and L. Y. Xiang, Synonym substitution-based steganographic algorithm with matrix coding, J. Chinese Comput. Syst., 36 (2015), 1296-1300.
    [16] H. H. Hu, X. Zuo, W. M. Zhang, et al., Adaptive text steganography by exploring statistical and linguistical distortion, IEEE Second International Conference on Data Science in Cyberspace, (2017), 145-150.
    [17] O. Russakovsky, J. Deng, H. Su, et al., Imagenet large scale visual recognition challenge, Int. J. Comput. Vision, 115 (2015), 211-252. doi: 10.1007/s11263-015-0816-y
    [18] L. Y. Xiang, X. B. Shen, J. H. Qin, et al., Discrete multi-graph hashing for large-scale visual search, Neural Process. Lett., 49 (2019), 1055-1069.
    [19] J. Wang, J. H. Qin, X. Y. Xiang, et al., CAPTCHA recognition based on deep convolutional neural network, Math. Biosci. Eng., 16 (2019), 5851-5861.
    [20] N. Kalchbrenner, E. Grefenstette and P. Blunsom, A convolutional neural network for modelling sentences, preprint, arXiv:1404.2188.
    [21] D. J. Zeng, Y. Dai, F. Li, et al., Aspect based sentiment analysis by a linguistically regularized CNN with gated mechanism, J. Intell. Fuzzy Syst., 36 (2019), 3971-3980. doi: 10.3233/JIFS-169958
    [22] R. H. Meng, S. G. Rice, J. Wang, et al., A fusion steganographic algorithm based on faster R-CNN, CMC-Comput. Mater. Con., 55 (2018), 001-016.
    [23] S. Q. Tan and B. Li, Stacked convolutional auto-encoders for steganalysis of digital images, Signal and Information Processing Association Summit and Conference, (2014), 1-4.
    [24] J. Q. Ni, J. Ye and Y. Yang, Deep learning hierarchical representations for image steganalysis, IEEE T. Inf. Foren. Sec., 12 (2017), 2545-2557.
    [25] Y. L. Qian, J. Dong, W. Wang, et al., Deep learning for steganalysis via convolutional neural networks, Proceed. SPIE, 9409 (2015), 94090J-94090J-10.
    [26] G. S. Xu, H. Z. Wu and Y. Q. Shi, Structural design of convolutional neural networks for steganalysis, IEEE Signal Proc. Let., 23 (2016), 708-712.
    [27] J. S. Zeng, S. Q. Tan, B. Li, et al., Large-scale jpeg image steganalysis using hybrid deep learning framework, IEEE T. Inf. Foren. Sec., 13 (2018), 1200-1214.
    [28] A. Krizhevsky, I. Sutskever and G. E. Hinton, Imagenet classification with deep convolutional neural networks, Proceedings of the 25th International Conference on Neural Information Processing Systems, 1 (2012), 1097-1105.
    [29] Y. Kim, Convolutional neural networks for sentence classification, Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, (2014), 1746-1751.
    [30] J. Turian, L. Ratinov and Y. Bengio, Word representations: A simple and general method for semisupervised learning, Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, (2010), 384-394.
    [31] G. E. Hinton, Learning distributed representations of concepts, Proceedings of the Eighth Annual Conference of the Cognitive Science Society, 1 (1986), 12-23.
    [32] Y. Bengio, H. Schwenk, J. Sencal, et al., Neural probabilistic language models, J. Mach. Learn. Res., 3 (2003), 1137-1155.
    [33] F. Morin and Y. Bengio, Hierarchical probabilistic neural network language model, Proceedings of the 10th International Workshop on Artificial Intelligence and Statistics, (2005), 246-252.
    [34] R. Collobert and J. Weston, A unified architecture for natural language processing: Deep neural networks with multitask learning, Proceedings of the 25th International Conference on Machine Learning, (2008), 160-167.
    [35] T. Mikolov, I. Sutskever, K. Chen, et al., Distributed representations of words and phrases and their compositionality, International Conference on Neural Information Processing Systems, (2013), 3111-3119.
    [36] B. Shen, C. W. Forstall, A. Rocha, et al., Practical text phylogeny for real-world settings, IEEE Access, 6 (2018), 41002-41012.
    [37] D. J. Zeng, Y. Dai, F. Li, et al., Adversarial learning for distant supervised relation extraction, CMC-Comput. Mater. Con., 55 (2018), 121-136.
    [38] R. Collobert, J. Weston, L. Bottou, et al., Natural language processing (almost) from scratch, J. Mach. Learn. Res., 12 (2011), 2493-2537.
    [39] Y. L. Boureau, N. L. Roux, F. Bach, et al., Ask the locals: Multi-way local pooling for image recognition, 2011 International Conference on Computer Vision, (2011), 2651-2658.
    [40] C. F. Yang, F. L. Liu, S. K. Ge, et al., Locating secret messages based on quantitative steganalysis, Math. Biosci. Eng., 16 (2019), 4908-4922.
    [41] C. F. Yang, X. Y. Luo, J. C. Lu, et al., Extracting hidden messages of MLSB steganography based on optimal stego subset, Sci. China Inform. Sci., 61 (2018), 119103:1-119103:3.
  • This article has been cited by:

    1. Mamta Keswani, Uttam Kumar Khedlekar, Optimal inventory management strategies for sequentially convertible items with different distribution patterns: A comprehensive stochastic approach, 2024, 0, 2155-3289, 0, 10.3934/naco.2024033
    2. Mamta Keswani, A comparative analysis of metaheuristic algorithms in interval-valued sustainable economic production quantity inventory models using center-radius optimization, 2024, 12, 27726622, 100508, 10.1016/j.dajour.2024.100508
    3. Mamta Keswani, 2024, chapter 11, 9798369356685, 283, 10.4018/979-8-3693-5668-5.ch011
    4. Xinyu Fu, Yanting Xu, The impact of digital technology on enterprise green innovation: quality or quantity?, 2024, 6, 2643-1092, 484, 10.3934/GF.2024019
    5. Mamta Keswani, Uttam Kumar Khedlekar, A fuzzy stochastic model incorporating advance sales, discounts, and carbon emission factors with comparative analysis of tuned meta-heuristic algorithms, 2024, 1432-7643, 10.1007/s00500-024-10375-0
    6. Mamta Keswani, Vinita Dwivedi, Uttam Kumar Khedlekar, Analyzing the impact of sustainability on EPQ model: Investigating price and sustainability level dependent demand using center-radius optimization techniques with Metaheuristic algorithms in interval valued environment, 2025, 0, 1547-5816, 0, 10.3934/jimo.2025001
    7. Mamta Keswani, 2024, chapter 9, 9798369375709, 127, 10.4018/979-8-3693-7570-9.ch009
    8. Fernando R. Mazarrón, José Álvarez, Jorge Cervera-Gascó, Jesús López-Santiago, Impact of Electricity Price Reductions on the Profitability and Sizing of Photovoltaic Systems in Agri-Food Industries, 2025, 15, 2077-0472, 631, 10.3390/agriculture15060631
    9. Silvia Carpitella, Joaquín Izquierdo, Trends in Sustainable Inventory Management Practices in Industry 4.0, 2025, 13, 2227-9717, 1131, 10.3390/pr13041131
    10. Vinita Dwivedi, Mamta Keswani, A dynamic pharmaceutical inventory investment management model during pandemics using metaheuristic algorithms, 2025, 27726622, 100570, 10.1016/j.dajour.2025.100570
    11. Abdi Shukri Yasin, Zsuzsanna Bacsi, Agritourism and Rural Development: A Global Bibliometric Analysis of the State of Research, Limitations, and Future Directions, 2025, 15, 2077-0472, 866, 10.3390/agriculture15080866
  • Reader Comments
  • © 2020 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(6552) PDF downloads(584) Cited by(57)

Figures and Tables

Figures(2)  /  Tables(3)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog