Trust in online media is increasingly compromised by deepfake multimedia, which undermines the authenticity of shared content. Existing detection techniques often perform well only on specific types of deepfakes, limiting their generalization ability and making them vulnerable in real-world applications. To address this, we propose a novel deepfake detection framework featuring an effective feature descriptor that integrates deep identity, behavioral, and geometric (DBaG) signatures, along with a classifier named DBaGNet. The DBaGNet classifier utilizes the extracted DBaG signatures and applies a triplet loss objective to enhance generalized representation learning for improved classification. These comprehensive DBaG signatures capture both facial geometry inconsistencies and behavioral cues, improving the detection of diverse deepfake types and enhancing generalization. We evaluate our approach using six benchmark deepfake datasets: WLDR, CelebDF, DFDC, FaceForensics++, DFD, and NVFAIR. Cross-dataset evaluations demonstrate significant performance gains over several state-of-the-art methods.

The rise of deepfakes poses a significant threat to public trust in digital content, particularly when presented on social media platforms where fabricated videos of political leaders, influencers, or celebrities can rapidly go viral. These convincing manipulations fuel misinformation, distort public opinion, promote scams, and challenge the integrity of content on social media. Additionally, with the recent shift from unimodal to multimodal deepfake generation, detection systems face increasing difficulty when identifying forgeries that combine both visual and audio manipulation. Traditional detection methods often fail to generalize across unseen manipulations or domains, limiting their practicality on dynamic, real-world social media content. To address this challenge, we propose domain-adaptive representation learning (DARL), a novel framework that leverages the inherent correlation between the audio and visual modalities using a deep canonical correlation analysis network. DARL effectively learns robust audio–visual representations by integrating knowledge from speech recognition domains, enhancing its ability to detect both LipSync and face-swap deepfakes. Our approach demonstrates improved accuracy and generalization across five benchmark datasets, along with a locally collected real-world deepfake advertisements dataset that features audio–visual deepfake social media content, highlighting its potential as a reliable defense mechanism against the spread of deepfake-induced misinformation on social media platforms.

Voice spoofing attacks pose a significant threat to automated speaker verification systems. Existing anti-spoofing methods often simulate specific attack types, such as synthetic or replay attacks. However, in real-world scenarios, the countermeasures are unaware of the generation schema of the attack, necessitating a unified solution. Current unified solutions struggle to detect spoofing artefacts, especially with recent spoofing mechanisms. For instance, the spoofing algorithms inject spectral or temporal anomalies, which are challenging to identify. To this end, we present a spectra-temporal fusion leveraging frame-level and utterance-level coefficients. We introduce a novel local spectral deviation coefficient (SDC) for frame-level inconsistencies and employ a bi-LSTM-based network for sequential temporal coefficients (STC), which capture utterance-level artifacts. Our spectra-temporal fusion strategy combines these coefficients, and an auto-encoder generates spectra-temporal deviated coefficients (STDC) to enhance robustness. Our proposed approach addresses multiple spoofing categories, including synthetic, replay, and partial deepfake attacks. Extensive evaluation on diverse datasets (ASVspoof2019, ASVspoof2021, VSDC, partial spoofs, and in-the-wild deepfakes) demonstrated its robustness for a wide range of voice applications.

The evolution of artificial intelligence (AI) techniques in recent years has increased the generation of fake content including AI-generated text, images, audio, and videos. Among which the fake visual content commonly known as deepfakes has imposed a great threat to society due to its negative impacts. To mitigate the adverse aspects of deepfakes, the research community has introduced various deepfakes detection methods. However, these deepfakes detection methods lack the interpretability and explainability of the decision-making process. The interpretable model increases trustworthiness as it provides the reasoning for classifying outcomes as real or fake. Therefore, in this paper, we have introduced ConvNext-PNet, which is a prototypical-based learning framework for the interpretable and explainable detection of visual deepfakes. In the proposed framework, prototype learning is incorporated into the modified ConvNext model that improves the discriminative features learning capability of the proposed framework along with the explainability aspect. The performance of ConvNext-PNet is evaluated on challenging datasets including FaceForensics++ (FF++), CelebDF, DFDC-P, and DeepFakeFace (DFF) datasets. The robustness of the proposed model is validated through various experiments along with the interpretability analysis. The quantitative results demonstrate the effectiveness of the model for the detection of visual manipulation, whereas the model interpretability and explainability aspect increases the trustworthiness via providing reasoning for the model predictions.

The proliferation of user-friendly deepfake creation tools poses a serious challenge, demanding robust and adaptable detection strategies. Existing approaches primarily focus on raw data analysis or identifying learned artifacts or manual data-driven rules resulting in the mis-classification of deepfakes with distorted facial poses. These architectures also neglect the potential power of combining learned visual features with explicit rules.To address this gap, we introduce RuleBoost, a novel NeuroSymbolic AI based framework that seamlessly fuses extracted visual features with automatically learned rules. Our framework employs a scalable rule-based learning approach to extract learned rules from facial geometry such as distance, area, and angle. The extracted rules integrated with deep visual features show promising results giving state-of-the-art area-under-the-curve of 96.19% and 95.44% on WLDR and FaceForensics++ Datasets respectively, surpassing other deep learning specific methods. To figure out the difference NeuroSymbolic approach makes, we also analyze the samples misclassified by traditional DL-based architectures and correctly classified by Rule- Boosted architecture. Based on empirical evidence, we conclude that DL-based architectures struggle to accurately detect real and fake samples when facial artifacts lead to poses that deviate from standard facial positioning, while RuleBoost exhibits improved performance in the same scenario.

Recently, we have observed an exponential growth in highly realistic deepfake videos, which are often used to spread disinformation, defame individuals, and even influence political outcomes. To combat these manipulated videos, researchers have proposed various deepfake detection techniques. Recent research has revealed that these detection techniques are vulnerable to different adversarial attacks. This paper examines the vulnerability of deepfake detectors to adversarial black-box attacks in terms of performing penetration testing to expose the existing defense benchmarks of current deepfake detectors. We present a perceptual facial mole black-box adversarial attack on deepfake detectors, where the attacker has limited knowledge of the architecture and settings of the detector. The proposed attack is visually natural and transferable based on the attention distraction mechanism, which distracts the model-shared attention patterns from the region of interest to other regions. We illustrate the efficacy of our attack on multiple cutting-edge deepfake detectors. This attack demonstrates that small perceptible perturbations that are visually natural on the facial face can disrupt and reduce the accuracy of the detectors significantly, up to 40.3%, with the highest success rate of 48.7%. Our findings highlight the necessity for proposing effective deepfake detectors that are resistant to black-box attacks.

Deepfake detection has become increasingly important in recent years owing to the widespread availability of deepfake generation technologies. Existing deepfake detection methods present two primary limitations; i.e., they are trained on a specific type of deepfake dataset, which renders them vulnerable to unseen deepfakes, and they regard deepfakes as a “black box” with limited explainability, making it difficult for non-AI experts to understand and trust the decisions. Hence, this article proposes a novel neurosymbolic deepfake detection framework that exploits the fact that human emotions cannot be imitated easily owing to their complex nature. We argue that deep fakes typically exhibit inter- or intra-modality inconsistencies in the emotional expressions of the person being manipulated. Thus, the proposed framework performs inter- and intra-modality reasoning on emotions extracted from audio and visual modalities using a psychological and arousal-valence model for deepfake detection. In addition to fake detection, the proposed framework provides textual explanations for its decisions. The results obtained using the Presidential Deepfakes Dataset and World Leaders Dataset of real and manipulated videos demonstrate the effectiveness of our approach in detecting deepfakes and highlight the potential of a neurosymbolic approach for expandability.

The rise of deepfake videos poses a serious threat to the authenticity of visual media, as they have a potential to manipulate public opinion, mislead individuals or groups, harm reputation, etc. Traditional methods for detecting deepfakes rely on deep learning models, which lack transparency and interpretability. To gain the confidence of forensic experts in AI-based deepfakes detector, we present a novel DFP-Net for detecting deepfakes using interpretable and explainable prototypes. Our method makes use of the power of prototype-based learning to generate a set of representative images that capture the essential features of genuine and deepfake images. These prototypes are then used to explain our model’s decision-making process and to provide insights into the features most relevant for deepfake detection. We then use these prototypes to train a classification model that can detect deepfakes accurately and with high interpretability. To further improve the interpretability of our method, we also utilize the Grad-CAM technique to generate heatmaps that highlight the regions of the image that contribute the most towards the decision of the model. These heatmaps can be used to explain the reasoning behind the model’s decision and provide insights into the visual cues that distinguish deepfakes from real images. Experimental results on a large-scale FaceForensics++, Celeb-DF and DFDC-P datasets demonstrate that our method achieves state-of-the-art performance in deepfakes detection. Moreover, the interpretability and explainability of our method make it more trustworthy to forensic experts by allowing them to understand how the model works and makes predictions.

The industrial demands of immersive videos for virtual reality/augmented reality applications are crescendo, where the video stream provides a choice to the user viewing object of interest with the illusion of “being there.” However, in industry 4.0, streaming of such huge-sized video over the network consumes a tremendous amount of bandwidth, where the users are only interested in specific regions of the immersive videos. Furthermore, for delivering full excitement videos and minimizing the bandwidth consumption, the automatic selection of the user’s Region of Interest in a 360° video is very challenging because of subjectivity and difference in contentment. To tackle these challenges, we employ two efficient convolutional neural networks for salient object detection and memorability computation in a unified framework to find the most prominent portion of a 360° video. The proposed system is four-fold: 1) preprocessing; 2) intelligent visual interest predictor; 3) final viewport selection; and 4) virtual camera steerer. First, an input 360° video frame is split into three Field of Views (FoVs), each with a viewing angle of 120°. Next, each FoV is passed to the object detection and memorability prediction model for visual interestingness computation. Furthermore, the FoV is supplied as a viewport, containing the most salient and memorable objects. Finally, a virtual camera steerer is designed using enriched salient features from YOLO and LSTM that are forwarded to the dense optical flow to follow the salient object inside the immersive video. Performance evaluation of the proposed system over our own collected data from various Websites as well as on public data sets indicates the effectiveness for diverse categories of 360° videos and helps in the minimization of the bandwidth usage, making it suitable for industry 4.0 applications.

The growing number of voice-enabled devices and applications consider automatic speaker verification (ASV) a fundamental component. However, maximum outreach for ASV in critical domains e.g., financial services and health care, is not possible unless we overcome security breaches caused by voice cloning algorithms and replayed audios. Therefore, to overcome these vulnerabilities, a secure ASV (SASV) system based on the novel sign modified acoustic local ternary pattern (sm-ALTP) features and asymmetric bagging-based classifier-ensemble with enhanced attack vector is presented. The proposed audio representation approach clusters the high and low frequency components in audio frames by normally distributing frequency components against a convex function. Then, the neighborhood statistics are applied to capture the user specific vocal tract information. The proposed SASV system simultaneously verifies the bonafide speakers and detects the voice cloning attack, cloning algorithm used to synthesize cloned audio (in the defined settings), and voice-replay attacks over the ASVspoof 2019 dataset. In addition, the proposed method detects the voice replay and cloned voice replay attacks over the VSDC dataset. Both the voice cloning algorithm detection and cloned-replay attack detection are novel concepts introduced in this paper. The voice cloning algorithm detection module determines the voice cloning algorithm used to generate the fake audios. Whereas, the cloned voice replay attack detection is performed to determine the SASV behavior when audio samples are simultaneously contemplated with cloning and replay artifacts.

Vehicular ad hoc networks (VANETs) were initially designed to assist in traffic management and delivery of safety messages. Due to the significant evolution in modern vehicles, the features offered by VANETs have expanded to include comfort and entertainment relevant services. This expansion has further increased the need to secure them. The security of VANETs is mainly dependent on sharing a cryptographic group key confidentially. Due to the frequent change in group membership, there is a need to update the group key repeatedly, which is difficult in highly dynamic networks like VANETs. Therefore, designing a secure, scalable, and efficient group key management protocol is challenging. Existing group key management protocols introduce a variety of limitations, including high computational cost for both group key computation and retrieval, additional computational and communication overhead when the membership in the group changes, and collusion among receiving vehicles. To overcome these limitations, this paper introduces a novel group key management protocol, ALMS. Performance analysis reveals that, compared to existing protocols, ALMS is more scalable since it introduces a low computational overhead for both the Trusted Authority (TA) and the receiving vehicles. Also, it does not suffer from the key distribution limitation as symmetric key management protocols do. Moreover, ALMS introduces only a light overhead on the TA for group membership change. This is achieved by decoupling the initialization from group key computation and performing it offline without affecting the size of the encrypted group key.

In the above article[1], on pages 11 and 14, math appears incorrectly in some paragraphs.

Abstract Clinical narratives such as progress summaries, lab reports, surgical reports, and other narrative texts contain key biomarkers about a patient's health. Evidence‐based preventive medicine needs accurate semantic and sentiment analysis to extract and classify medical features as the input to appropriate machine learning classifiers. However, the traditional approach of using single classifiers is limited by the need for dimensionality reduction techniques, statistical feature correlation, a faster learning rate, and the lack of consideration of the semantic relations among features. Hence, extracting semantic and sentiment‐based features from clinical text and combining multiple classifiers to create an ensemble intelligent system overcomes many limitations and provides a more robust prediction outcome. The selection of an appropriate approach and its interparameter dependency becomes key for the success of the ensemble method. This paper proposes a hybrid knowledge and ensemble learning framework for prediction of venous thromboembolism (VTE) diagnosis consisting of the following components: a VTE ontology, semantic extraction and sentiment assessment of risk factor framework, and an ensemble classifier. Therefore, a component‐based analysis approach was adopted for evaluation using a data set of 250 clinical narratives where knowledge and ensemble achieved the following results with and without semantic extraction and sentiment assessment of risk factor, respectively: a precision of 81.8% and 62.9%, a recall of 81.8% and 57.6%, an F measure of 81.8% and 53.8%, and a receiving operating characteristic of 80.1% and 58.5% in identifying cases of VTE.

Ruptured intracranial aneurysms are associated with a high rate of mortality and disability due to the difficulty in predicting the rupture and complexity of the condition itself. Clinical narratives such as progress summaries and radiological reports, etc. contain key biomarkers, medical signs, and symptoms. By applying ontology-based information extraction on clinical narratives to extract important evidences and subsequently using machine learning can help to make decision support tools for complex decision making such as prediction of aneurysm rupture. According to best of our knowledge, there doesn't exist any work to extract clinical features from clinical narratives to predict the rupture of intracranial/Brain aneurysms (BA). While no single factor individually contributes to the risk of rupture of a BA, it is important to consider the combined impact of these aspects to understand the rupture probability of the aneurysm. In this paper, we explore the impact of size as a relative factor in saccular aneurysms with respect to location, gender and symptomatic/asymptomatic aspects of BA. Our study involves descriptive and inferential statistical data analysis on features extracted from retrospective electronic health records (EHRs) using natural language processing (NLP) and ontology-based information extraction techniques. Our analysis shows that size alone is not the sole contributor for rupture but the combination of size, location and patient's gender can influence aneurysm ruptures. Our results also show interesting insight that on same vasculature location, the average size of ruptured aneurysms for females is always smaller than that of males.

BACKGROUND: Fulfilling the vision of Semantic Web requires an accurate data model for organizing knowledge and sharing common understanding of the domain. Fitting this description, ontologies are the cornerstones of Semantic Web and can be used to solve many problems of clinical information and biomedical engineering, such as word sense disambiguation, semantic similarity, question answering, ontology alignment, etc. Manual construction of ontology is labor intensive and requires domain experts and ontology engineers. To downsize the labor-intensive nature of ontology generation and minimize the need for domain experts, we present a novel automated ontology generation framework, Linked Open Data approach for Automatic Biomedical Ontology Generation (LOD-ABOG), which is empowered by Linked Open Data (LOD). LOD-ABOG performs concept extraction using knowledge base mainly UMLS and LOD, along with Natural Language Processing (NLP) operations; and applies relation extraction using LOD, Breadth first Search (BSF) graph method, and Freepal repository patterns. RESULTS: Our evaluation shows improved results in most of the tasks of ontology generation compared to those obtained by existing frameworks. We evaluated the performance of individual tasks (modules) of proposed framework using CDR and SemMedDB datasets. For concept extraction, evaluation shows an average F-measure of 58.12% for CDR corpus and 81.68% for SemMedDB; F-measure of 65.26% and 77.44% for biomedical taxonomic relation extraction using datasets of CDR and SemMedDB, respectively; and F-measure of 52.78% and 58.12% for biomedical non-taxonomic relation extraction using CDR corpus and SemMedDB, respectively. Additionally, the comparison with manually constructed baseline Alzheimer ontology shows F-measure of 72.48% in terms of concepts detection, 76.27% in relation extraction, and 83.28% in property extraction. Also, we compared our proposed framework with ontology-learning framework called "OntoGain" which shows that LOD-ABOG performs 14.76% better in terms of relation extraction. CONCLUSION: This paper has presented LOD-ABOG framework which shows that current LOD sources and technologies are a promising solution to automate the process of biomedical ontology generation and extract relations to a greater extent. In addition, unlike existing frameworks which require domain experts in ontology development process, the proposed approach requires involvement of them only for improvement purpose at the end of ontology life cycle.

Clinical notes contain information that is crucial for the diagnosis process. However, it is usually not properly manually analyzed due to the tremendous efforts and time it takes. Hence, an automated approach is eagerly needed to maximize clinical knowledge management and reduce cost. In this paper, we propose a framework SESARF: a Semantic Extractor to identify hidden risk factors in clinical notes and a Sentimental Analyzer to assess the severity levels associated with the identified Risk Factors. This tool can be customized to any disease using Linked Open Data (LOD) by selecting a specific disease and collecting its risk factors list from medical ontologies. The extracted knowledge can serve two purposes: 1) a feature vector is prepared, for any classifier in machine learning, containing risk factors and their weights based on our semantic enrichment and sentimental analyzer and 2) a proper comparison of the extracted information with wearable body sensors that can alert any major changes in a patient's health status to personalize treatment.

As more and more learners are opting for online learning, e-learning industry is working on improving learning experience of online user by providing relevant content and lot of additional references. Since online learners mostly prefer video tutorials, identifying major topics and subtopics covered in video tutorial is a big challenge. Recently, for efficient knowledge sharing and interoperability over web lot of attention is given to semantic web. In this paper, we propose a semantic web-based framework for automatic topic identification from video tutorials in order to identify the concepts and their associated semantically relevant resources. Our framework identifies relevant topic using disambiguation in e-learning resource which helps learners in more focused study.

Emerging healthcare systems are expected to leverage new Internet of Things (IoT) trends to enable preventive and personalized medicine. However, the success of such systems is entirely dependent on the ability to preserve patient privacy. This paper proposes a decentralized ontology based system architecture that caters to a healthcare organization's privacy needs as well as its enterprise security policy concerns considering futuristic IoT and Electronic Health Records (EHR) trends in healthcare. To identify privacy infringements and organizational policy violations across the institution, our proposed HealthCare Security and Privacy (HCSP) ontology is utilized in conjunction with three agent-level filters: the Exception Creation Agent (ECA), the Policy Mapping Agent (PMA), and the Privacy Assurance Filter (PAF). Preliminary evaluation has shown effectiveness of the HCSP ontology in terms of detection of policy compliance to HIPAA standards.

Tailoring treatment and clinical decision making to a person's unique characteristics is the next milestone for healthcare informatics, but for it to be accomplished, big data analytics for identifying risk factors and other hidden patterns among patients become paramount. In future these analytics will take the form of multicenter observational research, for which data preparation is vital. Specifically, quality data must be obtained in a timely manner while protecting the privacy of patients in the health records shared among researchers. Furthermore, the coordination and cooperation of a fluctuating number of medical data sources containing these records for clinical data distribution is an additional requirement in multicenter studies. Thus, we propose an autonomous decentralized, privacy-enabled data preparation architecture and novel SEDTM algorithm to meet these requirements, censuring sensitive information via filtration, and extracting relevant clinical data with a fully automated approach. Our evaluation demonstrates a 40% - 60% increase in the retrieval of quality patient data, compared to traditional semantic similarity, for our proposed SEDTM algorithm.

Ontology Based Information Extraction (OBIE) is being adopted in various domains in order to improve the system's precision and recall. Though use of multiple ontologies in different semantic based Information Extraction systems helps to improve the system extraction accuracy but the performance of system degrades significantly. This paper proposes autonomous decentralized kernel cache architecture to improve the query time for OBIE systems that utilize multiple ontologies. In order to minimize performance bottlenecks we propose caching most frequently used relations of the ontologies in the autonomous and decentralized kernel cache in order to reduce the query time against different semantic queries. Results show performance improvement for different queries made on YAGO2s and DBpedia ontologies using the proposed kernel based Onto-Cache.

The advent of social networking and open health web forums such as PatientsLikeMe, WebMD, ehealth forum etc. have provided avenues for social user data that can prove instrumental in suggesting futuristic trends in healthcare. Homophily in social networks is a vital contributor for analyzing patterns for medical conditions, diagnosis and treatment options. Since, members with similar medical issues contribute to a common discussion pool; this offers a rich source of information that can be utilized. This paper intends to explore growing trends in Mental Health and Behavioral Studies (MHB) which lays emphasis on co-existing conditions resulting in comorbidity. We present a novel approach where personality traits inferred from unstructured text of patients and general social users are compared via statistical analysis. This is achieved by our Psychiatric Disorder Determination (PDD) algorithm. Further, Social media data of users showing personality traits of patients is subjected to semantic based text classification using Natural Language Processing (NLP) and Ontology Based Information Extraction (OBIE) in our Addiction Category Determination (ACD) algorithm. This provides categorization of user journals to common topics of discussion by referring to ontologies DBpedia, Freebase and YAGO2s. The final category hence obtained can be predicted to be a trending subject of concern for users with Psychiatric disorders developing Addictive behavioral personalities.

The efficiency and effectiveness of the recovery of traceability links in requirements management is becoming increasingly important within interdisciplinary industry. Due to the complexity of production development such as automotive, software industry and the aerospace industry, managing requirements is indispensable and challenging. Products in these industries are constantly being updated and modified as the understanding of risks increases with experience and new products are developed in light of such risk. Therefore, the traceability links among the requirements artifacts, which fulfill business objectives, is so critical to reducing the risk and ensuring the success of products. To that end, this paper propose a semantic based traceability link recovery (STLR) architecture. According to best of our knowledge this is the first architectural approach that uses DBpedia knowledge-base and Bablenet 2.5 multilingual dictionary and semantic network for finding similarity among requirements and the automation of the recovery traceability links using our novel Triple extraction, and Triple disambiguation algorithm. Our preliminary results show the effectiveness in term of precision and recall compared to Vector Space Model and Wu Palmer algorithm.

Currently cyberspace has got about one billion registered websites, and it is imperative to accurately categorize voluminous number of website/URLs for the purpose of URL filtering and marketing segmentation. This paper presents autonomous decentralized semantic based large-scale URL/web classification system for web filtering using Yago2s and DS-onto knowledgebase. As many predefined categories are highly overlapping or semantically similar, proposed word sense disambiguation algorithm along with inference engine design brings high accuracy for classification of URLs in to 120 different categories. Evaluation results show that it achieves 90-93% of accuracy which is much higher than that obtained by currently used URL classification systems.

Low latency is the key requirement for cloud based web services particularly when there are heavy processes like inference from large knowledge-base are running and multiple simulataneous requests need to be entertained. This paper proposes Autonomous Decentralized L3 cache driver based URL filtering system, that uses two knowledge bases and proposal of L3 cache, to achieve low latency with high accuracy. The system creates ontology TDB files in way that most used relationships in ontology are in separate files. Thus the system caches the most frequently used relations of the ontology to minimize the time required by the inference engine of the text categorizer. Evaluation shows L3 cache achieved significant performance improvement in terms of processing time.

In the control of operation of the machinery and equipment of vehicles human error by operator makes accidents and disasters. In information technology system, there are also some accidents to system failure and data lost by the human error by operator. It is one of big issue for assurance system. To avoid human error by operator, emotion sense evaluation by Ontology rule is proposed. We configure this approach on Autonomous Decentralized Multi-Layered Cache system (ADMLC). We selected Galvanic Skin Response (GSR) value as the factor of human emotion for each operator. We try to sense operator unstable emotion by GSR value then request other operator to check whether his command order is true or not. We evaluated GSR value of nine operators by a week. Each operator has different GSR value and it is depending on person and environment. The analyzing method, Ontology rule is utilized in this proposal. Its rule engine always records the GSR value for ontology rule engine. The dynamic value of GSR shows operator's emotion in real time. Ontology rule engine decision makes the result of GSR value using the historical data. The result of emotion sense accuracy shows more than 90 % by evaluation. By the Ontology emotion sense rule engine on ADMLC system, it shows as one of possibility of method to avoid the system failure or data lost by the human error.

Community Context-attribute-centric Collaborative Information Environment (CCCIE) enables provision of services to users with same interests, by employing time awareness and demand-oriented perspective to transform one to one (1-1), one dimensional location based service paradigm to one to many (1-N), n-dimensional situation-aware community services. Unlike conventional Location Based Services (LBS) which involve selection of most suitable service either by manipulating location context at centralized server, or context-aware selection among all discovered services at service-consumer logic (at end node), the circumspect design of Autonomous Decentralized Community System (ADCS)framework empowers the transparent access of right service to right users at right place and right time, in highly dynamic ubiquitous environment by collaborative decentralized processing of various surrounding contexts like traffic and weather conditions, topographic features of towns and service validity time etc.

Recent headway in the domains of mobile communication and ubiquitous computing has given surge of interest to mobile commerce applications. Current information systems promote concept of information service provision to anyone, anytime and anywhere. Autonomous Decentralized Community System has been proposed to meet the varying requirements of users. In this paper, the decentralized search service is taken up as an application. This retrieves service in which the service offer end time has been decided beforehand. Needs of this application are the accuracy, the adaptability, and the improvement of the assurance is requested. There are two requirements in this service. First, it is to prevent the network being crowded with the request message from the node where SP that can enjoy serving doesn't exist. And, it is to prevent the stability of the entire system from decreasing to update cash information. To solve these, it proposed the Autonomous Construction Technology and the Autonomous Change Adjustment Technology. Simulation results demonstrate the effectiveness of proposed technology in terms of improvement in service assurance and confirms the trade-off between accuracy and system stability.

Community time-distance-oriented context-aware information environment (CTCIE) incorporates time-awareness and demand-oriented perspectives to revamp one-dimensional rudimentary location based services (LBS) into three dimensional community pervasive services (CPS). The proposed Autonomous Decentralized Community System (ADCS) framework enables provision of customized CPS to specific users in specific place at specific time. ADCS employs novel idea of ripple-based multi-target context-cognizant service discovery (RMCSD) technique to empower flexible means to sophisticated selection of right service to right user at right place in transparent way, by utilizing context information of surrounding situations (traffic conditions, topographical features, service validity time etc). To further enhance timeliness, hybrid pull-push mechanism of RMCSD is extended to cache the service contents of the most relevant service. However, given multiple service providers, there exists tradeoff between accuracy of information and overall response time perceived by the users. The vigilant design of proposed autonomous community service area reconstruction technology (ACS-ART) procures high assurance: provision of right service with least access time. Simulation results verify the effectiveness of the proposed techniques.

In recent years, with development in wireless communication technologies and sensor devices, wireless sensor networks have gained worldwide attention. In production plant monitoring systems, especially food factory monitoring systems, it is needed to relocate sensors in accordance with reorganization of production lines. A conventional centralized management systems isn't able to cope with huge number of sensors and rapidly changing topology. Autonomous Decentralized Community Wireless Sensor Network System (ADCWSN) is presented to solve these problems. In ADCWSN, all nodes are autonomous, and each node copes with problems by mutual corporation according to the situation. (This set of nodes which mutual corporate forms community.) In this paper, Autonomous Initialization Technology and Autonomous Community Collaboration Technology are proposed to satisfy expandability and the effectiveness of proposed technologies is shown through simulation.

In today's aggressive and competitive marketplace, there is substantial need for corporations to incorporate service level agreement (SLA) in their enterprise; and keep user differentiation with respect to information volume, response time, or system availability; to make their business more idiosyncratic. Faded information field (FIF) was proposed to deliver assurance in distributed information system but it does not fulfil requirement of user differentiation to attain heterogeneous service levels. Due to structural characteristics of FIF, SLA-based mobile agent navigation technology is proposed, which is composed of autonomous node, response time monitoring, local status notification between neighbor nodes, and finally the Pull-MA navigation decision based on locality statistics information. As a result, the requests are able to avoid local congestion zones without a centralized dispatcher or workload manager, leading to globally accomplishing the heterogeneous service level objectives of each user even in case of subsystem failure and ever-changing environment. The effectiveness of the proposed technology has been proved through simulation, and the results show that the navigation technology increases an average of 60% the service satisfaction ratio. Since proposed navigation of Pull-MA is highly dependent on value of maximum access time (MAT), MAT per node was simulated for different techniques. Moreover, response time of all nodes in the network was simulated and satisfaction ratio for various possible techniques, along with that proposed has been shown for comparison