Smart contracts represent a predefined set of rules invoked when specific conditions are met within blockchain networks, eliminating the need for centralized authority to validate transactions. The absence of central authority can potentially expose smart contracts to fraudulent behavior. Moreover, implementation flaws in smart contracts can be exploited to cause unintended behavior, resulting in security or financial risks. Traditionally, the identification of vulnerabilities in smart contracts has relied on methods such as pattern matching, data flow analysis, and input testing. While these techniques are foundational, they are constrained by human limitations and may not comprehensively address the full spectrum of potential issues. This necessitates more advanced approaches to ensure robust security and reliability. Therefore, in the literature, numerous researchers have leveraged different Machine Learning (ML) and Deep Learning (DL) techniques to classify normal and malicious smart contracts. However, existing literature either grapples with class imbalance issues or relies on conventional methods. Moreover, existing research often falls short of locating the exact location of malicious code within the smart contracts. Therefore, to address these gaps, this paper proposes a novel model called the Dual-Branch Encoder Siamese Network (DBESN) for detecting malicious smart contracts. Furthermore, this model is extended to precisely identify the region of the vulnerable code fragment within the smart contract using the Local Interpretable Model-Agnostic Explanations (LIME) algorithm. Experimental results demonstrated a performance Accuracy of 98.62% and 99.30% F1-Score with an inference time of 0.296 seconds. Given the high performance coupled with the low inference time of the proposed DBESN model, it is suitable for deployment within blockchain networks to detect and identify malicious smart contracts effectively and efficiently.

Recommender systems (RSs) use information filtering to recommend information of interest (to a user). Similarly, personalisation can be adopted for recommendations in e-market. We propose a new and innovative system called as interest-based recommender system (IBRS) for personalisation of recommendations. The IBRS is an agent-based RS that takes into account user's preferences. It can transform a standard product (or service) into a dedicated solution for an individual. The system discovers interesting product alternatives based on user's underlying mental attitudes (likes and dislikes) during the repair process using argumentation. The proposed method combines a hybrid RS approach with automated argumentation-based reasoning between agents. The system improves results by improving the recommendation repair activity. We consider a book recommendation application, for experiment to carry out the system's (objective and subjective) evaluation using standard metrics. The experiments confirm that the proposed IBRS improves user's acceptance of the product as compared with a traditional hybrid method and an argumentation-enabled state-of-the-art recommendation method. The system has been found to be more effective than its traditional counterpart when dealing with the new user problems.

Feature extraction and classification based on feature sets are two major components of steganalysis process. The high dimension of feature sets used for steganalysis makes classification a complex and time-consuming process. This paper proposes a novel feature selection algorithm (FS-SDS) for steganalysis. FS-SDS is a wrapper-type feature selection algorithm which selects reduced feature set using Stochastic Diffusion Search. The Stochastic Diffusion Search is a generic population-based search method, which has been adopted successfully in this work for steganalytic feature selection. The experiments are conducted with steganograms of the common JPEG steganography techniques. To show the usefulness and effectiveness of FS-SDS, experiments were conducted on two different feature sets used for steganalysis. The experimental results show that the proposed feature selection not only effectively reduces the dimensionality of the features, but also improves the detection accuracy of the steganalysis process.

In distributed, heterogeneous and network-connected collaborative environments where resources are provided to diverse unknown users for their applications, it is necessary to define access control for resources. Access control for such systems is defined as the ability to authorise or repudiate access to resources by a particular user. Traditional access control solutions are inherently inadequate for collaborative systems because they are effective only in situations where the system knows in advance which users are going to access the resources and what are their access rights so that they can be predefined by the developers or security administrators, but in collaborative systems the number of users as well as their usage on resources is not static. Targeting collaborative systems, a fine grained, flexible, persistent trust-based model for protecting the access and usage of digital resources is defined in this paper using radial basis function neural network (RBFNN). RBFNN classifies the users requesting the resources as trustworthy and non-trustworthy based on their attributes. RBFNN is used for classification because of its ability to generalise well for even unseen data and non-iterative method employed in its training. A proof of concept implementation backed by extensive set of tests on the real data collected for one such collaborative systems, i.e. Enabling Grids for E-Science grid demonstrated that the design is sound for collaborative systems where access of resources are provided to large and unknown users with their variant set of requirements.

This paper presents a multi- agent system architecture for intelligent watermarking for securing fingerprint images. The proposed watermarking method uses a fuzzy-PSO based hybrid approach to secure a person's fingerprint image by watermarking it with its corresponding face image. As fingerprint databases are large, processing huge image data in real time is difficult. The proposed work uses a multi-agent system as a distributed system for performing watermarking of fingerprint images, where various subtasks are performed in parallel in the distributed system. For watermarking input fingerprints, each image is divided into blocks and type-2 fuzzy logic is used to calculate the watermarking strength of each block based on its features. Particle Swarm Optimization (PSO) is used to find optimum DCT coefficients of the image block to be used for watermark embedding in such a way that the quality and minutia matching ability of the host fingerprint are preserved. The proposed system WoFMAS (Watermarking of Fingerprints using Multi-Agent System) has multiple agents working on different blocks of the input fingerprint concurrently for efficiently distributing data and gathering results. The experimental study is done on the FVC 2004 fingerprint database and the results show that our hybrid approach gives better results in terms of watermarked image quality and robustness than other fuzzy based and other PSO based approaches in the literature.

Recommender Systems are a class of applications which are used to overcome the problem of information overload. They use the opinions of members of a community to help individuals in that community identify the information most likely to be interesting to them or relevant to their needs, by drawing on user preferences and filtering the set of possible options to a more manageable subset. The key element of such user-support systems is the user model. Traditional techniques used to create user models are usually too rigid to capture the inherent uncertainty of human behavior. Fuzzy sets can handle and process uncertainty in human decision-making and if used in user modeling can be of advantage as it will result in recommendations closely meeting user preferences. In this paper, a hybrid multi-agent recommender system is designed and developed where user's preferences; needs and satisfaction are modeled using interval type-2 (IT2) fuzzy sets. This results in improving the prediction accuracy of the system and hence better recommendations are generated. Experimental study was conducted on book recommender system and promising results were obtained.

Software agents are dynamic, autonomous, reactive, pro-active, possibly persistent and sociable software entities that are designed to achieve their desired goals by enabling personalized solution for complex systems. Agent mediated systems provide the necessary decision making support to facilitate the automation of processes, especially in dynamic environments like e-market. This paper proposes and simulates framework of a distributed multi-agent e-market environment comprising of self-interested buyers and sellers where buyers and sellers communicate through asynchronous message passing. The proposed architecture is situated in a single cycle, sealed bid reverse auction scenario, where self-interested buyers and sellers compute reputation of participants dynamically by utilizing both web and agent technologies. It enables a participating agent to compute the trustworthiness of other agents dynamically based on its past experience and other changing parameters of e-market.

This paper proposes a novel blind Steganalysis process, for colored JPEG images. Extreme Learning Machine (ELM) has been used in the paper to classify the images into stego images and non-stego images. The feature set used for classification of images consists of 810 features. First 405 features are based on Markov random process applied on correlations among JPEG coefficients of image. Calibration is applied on these Markov features to get the remaining 405 features. These calibrated features are the difference between the Markov features of the image and Markov features of a reference image, obtained by decompressing, cropping and recompressing the image. Experimental results show that our proposed ELM based steganalysis method clearly outperforms other SVM based steganalysis methods in terms of percentage of correctly classified images and in terms of time taken for both training and testing. The fast speed of the proposed method due to fast learning time of ELM makes it useful for real-time steganalysis.

SUMMARY Present day sophisticated and innovative attacks have resulted in exponentially increasing security problems. This paper therefore presents a three‐phased threat‐oriented security model to meet the above security challenges as a part of proactive risk management. This model is based on a spiral process for software development because it is a risk driven approach and provides an incremental method for a progressively growing system with decreasing risk. Integration of threat management during the development process in the proposed work provides necessary security cover against both unforeseen and known threats. Identification of these threats has been made possible by fusion of a threat modeling process and research honeytokens in conjunction with a statistical model in the first phase. Necessary security measures to mitigate the above identified threats have been adopted in the second phase using multiagent system planning. Risk reduction as a result of adoption of countermeasures has been evaluated in the third phase using meta‐agents in association with fuzzy logic in a multiagent environment. The proposed proactive measures of this model have been demonstrated with a case study on ‘Online Banking’ to show its feasibility and has been implemented using Java Agent Development Environment, Apache Tomcat Server, with MySQL Server at the backend. Copyright © 2012 John Wiley & Sons, Ltd.

The BPEL specification focuses on business processes, the activities of which are assumed to be interactions with Web services. These multiple services are called composite Web service. WSDL is the W3C specification for developing a Web service. However, change in the business requirements over a period of time result in multiple versions of Web services. Maintaining multiple versions is difficult especially for organizations, with established products and a large customer base. The problem aggregates further when these Web services are bundled into a business process. In our earlier work, we have extended WSDL to WSDL-Temporal (WSDL-T) that eases the management of various versions of a Web service. WSDL-T allows access to a particular version of an operation of a Web service from the client. But, existing specifications of the Bussiness Process Execution Language (BPEL) does not support the new artifacts proposed in the WSDL-T. In the presented work, the BPEL specifications are extended to BPEL-Temporal. BPEL-Temporal has introduced a new artifact that allows accessing a particular version of an operation from the bunch of available versions within a single Web service. BPEL-Temporal enables the BPEL processes to call a new or old version of the constituent Web service dynamically. The BPEL-T helps in easy and better management of business processes.

This paper presents a hybrid intelligent watermarking scheme for securing fingerprint images in the wavelet domain. The proposed method uses an NN-PSO based hybrid approach to secure a person's fingerprint image by watermarking it with its corresponding face image and demographic information. The input fingerprint image is divided into blocks in the proposed approach and a feed forward neural network is used to calculate the number of bits to be embedded in each image block based on its features. PSO is used to find optimum wavelet coefficients of the image block to be used for watermark embedding in such a way that the quality and minutia matching ability of the host fingerprint are preserved. The experimental study is done on the FVC 2004 fingerprint database and the results show that our hybrid approach gives better results (quality and robustness) than other neural network based and PSO based approaches in the wavelet domain. As the host fingerprint is watermarked with other biometric information of the same person, the proposed work finds application in security implementations based on multimodal biometric authentication.

The proactive recommender system automatically delivers (i.e. pushes) recommendations to the user, without explicit request from him. The push model seems to be very effective in the applications where the availability of items changes often and rapidly, as it helps users timely receive their interested information. However, if the system pushes uninterested information to the user, or even pushes interested information to the user but at inappropriate context, then the user's acceptance of proactively delivered recommendations will decrease enormously. Hence for improving user's acceptance in proactive recommender systems, determination of right context (situation assessment) and finding relevant items for the target user are very crucial. This paper presents a Situation-Aw are Proactive Recommender System (SAPRS) that pushes relevant items to the target user at the right context only. The recommendation process in the proposed system is divided into two phases: (i) situation assessment phase and the (ii) item assessment phase. In situation assessment phase, the SAPRS system analyzes the current situation i.e. whether or not the current context needs a recommendation. In item assessment phase, the suitable items are selected as recommendations using a collaborative filtering approach. SAPRS uses fuzzy logic as an inference technique to handle uncertainty within situation assessment phase. The prototype of SAPRS has been designed and developed for restaurant recommendations. Performance of the implemented prototype system is evaluated using users' subjective feedback.

The purpose of open, distributed systems, is to enable coordinated resource sharing to diverse users. In order to protect each participant's privilege and security, a secure and efficient access control is essential. Due to limited or lack of knowledge about user's identities in open systems, access control need to be specified in terms of the attributes and properties of the users. Moreover, the attribute based access control methods are known for their flexibility and dynamicity. This paper puts forward a novel attribute based trust model using Radial Basis Neural Network to define the access control. RBFNN is used because of its ability to generalize well even for unseen data and fast learning data. The trustworthiness of the requestors is computed using RBFNN, maintaining the continuity of access decisions. The framework is validated on the real data of EGEE grid, a distributed system and found to be performed better than feed forward neural network.

Advancements in wireless technology, has led to the increased use of wireless handheld mobile devices. The proliferation of a vast number of these devices and the curiosity of the young generation about the information available on the internet anytime anywhere has resulted in a ubiquitous computing scenario. Ubiquitous computing consists of various kinds of computational devices, networks, collaborating software and hardware entities that are required to be interoperable and discoverable in order to communicate with each other. Interoperability is the ability of two or more systems or components to exchange information and to use the information that has been exchanged. In a highly dynamic and distributed environment like MANets, the basic issue is the discoverability of devices and services by the other entities in the system. This paper presents a framework that supports the discoverability and interoperability of the devices forming an ad-hoc network for ubiquitous applications. The proposed approach uses the Mobile Process (Agents) Group to allow devices on the network to interact with one another. It uses group communication semantics to discover devices and interoperate with those devices. The approach does not impose any constraints at device development time instead uses only run-time constraints.

Explosive growth of data on the web demand techniques, which would enable the user to access desired information. In Information retrieval Document Classification is prerequisite. In practice many classification techniques were and are in use. Term Frequency-Inverse Document Frequency (TF-IDF) is an approach which represents documents based on the frequency of terms in documents. Limitation of this approach is high dimensionality of data. Moreover it does not consider the relations among the terms, resulting in less precise and noisy end result. In our approach we are using weighted Concept Frequency-Inverse Document Frequency (CF-IDF) with background knowledge of domain Ontology, for classification of RSS feed News Items. Metadata information of news items has been used to assign weight to the identified concepts. No trained classifiers are required as Ontology itself acts as a classifier. We have designed ontology based on news industry standards. This classification approach considers relations among the concepts and properties. It results in reduction of noise in final output. It considers only the key concepts of a domain for classification instead of all the terms, which curbs the problem of dimensionality. Evaluation of experimental results reveals that proposed approach gives better classification results.

Nowadays, companies recognize the need to be customer driven by providing superior service to satisfy customers' needs. But as customers and their needs grow increasing diverse, unnecessary cost and complexity are inevitably added to operations. Service providers discovered the new frontier in business competition: “Collaborative Customization.” This approach follows three steps: first to conduct a dialogue with individual customers to help them articulate their needs; second, to identify the precise offering that fulfills those needs; and third, to make customized products for them. Web services deployed over the Web are accessible to a wider user base. Web services are designed and contracted to meet the need of large number of users. Many a times, multiple customizations of the base functionality is required to cater the need of multiple set of users. This forces service provider to deploy multiple Web services customized for each set of users, which results in increasing cost of infrastructure and maintenance. Since, multiple versions of customized Web services are deployed multiple times at different URLs, it is difficult and costlier to maintain, update and backup these services and their data. The objective of this work is to reduce the efforts and cost that resulted due to these multiple versions of the Web Services. We have extended WSDL and WSDL-T to WSDL-TC that aims at reducing the cost by maintaining the different collaborative customized versions of operations of the Web service in a single deployment. The approach also manages access control of these operations to their respective groups. WSDL-TC being an extension of WSDL-T is capable of managing versions of each customized operation that resulted due to the changes in their business requirements over a period of time. The paper presents an example of hotel reservation web service to present WSDL-TC approach. WSDL-TC also eases the task of web service administrators as they have to manage the single instance instead of multiple instances of a Web service. Moreover, WSDL-TC based services when deployed in the cloud environment may help in achieving greater degree of multi-tenancy further reducing the cost for service producers.

Collaborative Filtering is one of the most researched techniques. It generates recommendations from similar taste users in a group. In this paper, Information Theoretic Techniques are used to propose an Online Recommendation Generator based on Collaborative Filtering. It initially generates preliminary recommendations based on positive and negative user preferences and further refines these preliminary recommendations based on opposite user preferences. Experiments are conducted using MovieLens Dataset and considerable improvement in accuracy is seen in the results.

In bug repositories, we receive a large number of bug reports on daily basis. Managing such a large repository is a challenging job. Priority of a bug tells that how important and urgent it is for us to fix. Priority of a bug can be classified into 5 levels from PI to P5 where PI is the highest and P5 is the lowest priority. Correct prioritization of bugs helps in bug fix scheduling/assignment and resource allocation. Failure of this will result in delay of resolving important bugs. This requires a bug prediction system which can predict the priority of a newly reported bug. Cross project validation is also an important concern in empirical software engineering where we train classifier on one project and test it for prediction on other projects. In the available literature, we found very few papers for bug priority prediction and none of them dealt with cross project validation. In this paper, we have evaluated the performance of different machine learning techniques namely Support Vector Machine (SVM), Naive Bayes (NB), K-Nearest Neighbors (KNN) and Neural Network (NNet) in predicting the priority of the newly coming reports on the basis of different performance measures. We performed cross project validation for 76 cases of five data sets of open office and eclipse projects. The accuracy of different machine learning techniques in predicting the priority of a reported bug within and across project is found above 70% except Naive Bayes technique.

Opinions elicited from various groups of society (stakeholders) need to be incorporated into developing socially aware e-learning courses. Elicitation of opinion in initial phases of the designing process helps in chalking out needs of the stakeholders and their expectations from the course. This paper proposes a Social Opinion Framework to involve stakeholders in pre-design as well as post-deployment stages of e-course development. The framework suggests identification of stakeholders from the relevant domains and analysis of their roles. The respective opinions are to be elicited, prioritized and incorporated next in order to modify the course. The paper also discusses how the respective contributions can be instrumental in providing necessary support structures and quality parameters to improve upon the course composition, thus providing a better e-learning experience to the learners.

Learning resources which were earlier limited to libraries only are available on the Web as well these days. These learning resources can be in the form of eBooks, eJournals, experts' blogs, and even a general communication over a topic among the community can be used as an important source of information. Retrieving relevant informative resources from the huge voluminous Web and, organizing them for later use is a tedious and time consuming job. This paper proposes a Multi Agent System for automatic retrieval and organization of the Web content for learning purposes. The system is designed to provide the relevant and useful content from the Web to a learner at any time and, hence can be used for informal e-Mentoring, a sub-task of e-Learning. A prototype of the system is implemented using JADE.

The goal of Recommender Systems (RSs) is to help users to deal with the problem of information overload by facilitating access to relevant items that are valuable to them. If the recommended items match the user preferences, user trust in the system increases and the user start liking the system and uses it more frequently. Trust enabled Argumentation Based Recommender System (TABRS) designed and developed in this paper recommends items of interest to the user by using a hybrid approach for recommendation. These recommendations are further improved using argumentation to convince users about the product. TABRS is an agent-based recommender system that takes into account user's changing preferences to generate interesting recommendations. TABRS combines hybrid recommender system with automated argumentation between agents. The system also improves recommendation repair activity by discovering interesting alternatives based on user's underlying mental attitudes. We implemented the system using Jason for building agents enabled with inference and interaction capabilities. The experimental study is conducted for a Book Recommender System and performance of the proposed system is evaluated using precision and recall metrics.

Presented is an efficient watermarking scheme using Particle Swarm Optimization (PSO) to watermark host fingerprint images with their corresponding facial images in the Discrete Cosine Transform (DCT) domain. PSO is used to find the best DCT coefficients’ locations in the host image where the facial image data can be embedded, so that the distortion produced in the host image is minimum. The objective function for PSO is formulated in terms of the Structural Similarity Index (SSIM) and the Orientation Certainty Level Index (OCL) so as to base it on the simple visual effect of the human visual perception capability and correct minutia prediction ability. The results exhibit better watermarked image quality while retaining the feature set of the original fingerprint. Moreover, the proposed technique is robust so that the extraction of watermark is possible even after the watermarked image is exposed to attacks. As a result, at the receiver’s end, the watermarked fingerprint image and the extracted facial image can be verified for a secure and accurate biometric based personal authentication.

Conformance to social context while designing an e-learning course is crucial in enhancing acceptability of the course. Building socially aware e-learning courses requires elicitation of social opinion from various stakeholders associated with the system. Stakeholders are disparate in their perception towards the intricacies of the system, leading to generation of numerous assorted ideas. Knowledge Management (KM) assimilates these ideas to bring congruency into the system. This paper proposes i) a model KMeLS (Knowledge Management in e-Learning Systems) built upon the SECI (Socialization, Externalization, Combination and Internalization) framework, and ii) an algorithm PARSeL (Prioritizing Alternatives using Recommendations of Stakeholders in e-Learning) to incorporate KM into designing an e-learning course. PARSeL prioritizes the content using stakeholder recommendations using Analytic Hierarchy Process (AHP) and fuzzy modeling. A case study is also presented with a goal of prioritizing a set of programming languages for an online computing course. The proposed methodology can be promising in recommending appropriate content for the e-learners and can be implemented to benefit e-learning organizations in a wider spectrum.

There were times when software systems and networks posed no or very little security problems. However, with expanding connectivity during last few years problem of security has been making headlines. This is due to increase in threat environment and breach of security vital to the interest of end users. Keeping in view the security requirements in the present system all the attack points which can be threatened have to be identified, analyzed and remedial measures taken at the initial stage of software development process. The use of multiple techniques is the subject of research for deriving security requirements. In this paper, we are overlapping misuse case and attack trees techniques to propose a new technique named "Hybrid Technique". This Hybrid Technique merges the strengths of misuse cases and attack trees making the system stronger to mitigate weaknesses effectively in large and complex systems. In our approach we firstly identify threats using the concepts of threat modeling, and then map these threats into security requirements using Hybrid Technique. In the case study, we have used this technique for specifying security requirements for wireless hotspots.