Offline writer identification poses significant challenges when training data is scarce, and handwriting styles exhibit high intra-writer variability. This scenario is common in practical applications such as forensic analysis and historical document authentication, where only a limited number of handwritten samples are available per writer. In this paper, we explore the viability of using diffusion models to capture writer-specific traits under such challenging conditions. Specifically, we investigate their performance in both text-dependent and text-independent setups, where lexical similarity varies across samples. We propose a novel diffusion-based writer identification framework that integrates a style encoder and handcrafted textural features in a joint training pipeline. Our approach is evaluated on a recent dataset with high intra-writer variability as well as three benchmark datasets (IAM, CERUG-EN, and CVL). Experimental results demonstrate that while diffusion models excel in text-dependent scenarios, their generalization capability diminishes in text-independent settings due to the entanglement of content and style features. This study highlights both the promise and the current limitations of generative diffusion models for fine-grained handwriting style modeling. We identify avenues for improving generalization through disentangled representations, domain adaptation, and hybrid discriminative-generative architectures. The proposed framework contributes to the growing efforts toward scalable, style-aware writer identification in real-world, unconstrained handwriting scenarios.

Quality-of-Service (QoS) prediction is a critical task in the service lifecycle, enabling precise and adaptive service recommendations by anticipating performance variations over time in response to evolving network uncertainties and user preferences. However, contemporary QoS prediction methods frequently encounter data sparsity and cold-start issues, which hinder accurate QoS predictions and limit the ability to capture diverse user preferences. Additionally, these methods often assume QoS data reliability, neglecting potential credibility issues such as outliers and the presence of greysheep users and services with atypical invocation patterns. Furthermore, traditional approaches fail to leverage diverse features, including domain-specific knowledge and complex higher-order patterns, essential for accurate QoS predictions. In this paper, we introduce a real-time, trust-aware framework for temporal QoS prediction to address the aforementioned challenges, featuring an end-to- end deep architecture called the Hypergraph Convoluted Transformer Network (HCTN). HCTN combines a hypergraph structure with graph convolution over hyper-edges to effectively address high-sparsity issues by capturing complex, high-order correlations. Complementing this, the transformer network utilizes multi-head attention along with parallel 1D convolutional layers and fully connected dense blocks to capture both fine-grained and coarse-grained dynamic patterns. Additionally, our approach includes a sparsity-resilient solution for detecting greysheep users and services, incorporating their unique characteristics to improve prediction accuracy. Trained with a robust loss function resistant to outliers, HCTN demonstrated state-of-the-art performance on the large-scale WSDREAM-2 datasets for response time and throughput.

The rapid advancement of generative models and their misuse have made deepfake detection a crucial area of research. However, existing datasets and detection techniques predominantly focus on frontal-face perspectives, leaving sideface views largely underexplored. To bridge this gap, we present IndicSideFace, a novel dataset specifically curated for advancing deepfake detection on side-face perspectives of Indian subjects. This dataset encompasses a diverse range of side-face angles, varying lighting conditions, and demographic attributes, providing a comprehensive benchmark for evaluating detection algorithms. Our experiments using state-of-the-art models highlight the unique challenges posed by side-face deepfakes, such as partial facial feature visibility and uncommon head poses. The findings reveal significant limitations in existing detection approaches when applied to side-face perspectives, underscoring the need for specialized solutions. With IndicSideFace, we aim to strengthen the resilience of deepfake detectors and stimulate further research in this critical yet underexplored domain.

In the film industry, movie posters have been an essential part of advertising and marketing for many decades and continue to play a vital role even today in the form of digital posters through online, social media, and over-the-top (OTT) platforms. Typically, movie posters can effectively promote and communicate the essence of a film, such as its genre, visual style/tone, vibe, and storyline cue/theme, which are essential to attract potential viewers. Identifying the genres of a movie often has significant practical applications in recommending the film to target audiences. Previous studies on genre identification have primarily focused on sources such as plot synopses, subtitles, metadata, movie scenes, and trailer videos; however, posters precede the availability of these sources and provide prerelease implicit information to generate mass interest. In this article, we work for automated multilabel movie genre identification only from poster images, without any aid of additional textual/metadata/video information about movies, which is one of the earliest attempts of its kind. Here, we present a deep transformer network with a probabilistic module to identify the movie genres exclusively from the poster. For experiments, we procured 13882 number of posters of 13 genres from the Internet movie database (IMDb), where our model performances were encouraging and even outperformed some major contemporary architectures.

Optimizing tail latency while efficiently managing computational resources is crucial for delivering high-performance, latency-sensitive services in edge computing. Emerging applications, such as augmented reality, require low-latency computing services with high reliability on user devices, which often have limited computational capabilities. Consequently, these devices depend on nearby edge servers for processing. However, inherent uncertainties in network and computation latencies—stemming from variability in wireless networks and fluctuating server loads—make service delivery on time challenging. Existing approaches often focus on optimizing median latency but fall short of addressing the specific challenges of tail latency in edge environments, particularly under uncertain network and computational conditions. Although some methods do address tail latency, they typically rely on fixed or excessive redundancy and lack adaptability to dynamic network conditions, often being designed for cloud environments rather than the unique demands of edge computing. In this paper, we introduce SafeTail, a framework that meets both median and tail response time targets, with tail latency defined as latency beyond the percentile threshold. SafeTail addresses this challenge by selectively replicating services across multiple edge servers to meet target latencies. SafeTail employs a reward-based deep learning framework to learn optimal placement strategies, balancing the need to achieve target latencies with minimizing additional resource usage. Through trace-driven simulations, SafeTail demonstrated near-optimal performance and outperformed most baseline strategies across three diverse services.

In the realm of modern service-oriented architecture, ensuring Quality of Service (QoS) is of paramount importance. The ability to predict QoS values in advance empowers users to make informed decisions, ensuring that the chosen service aligns with their expectations. This harmonizes seamlessly with the core objective of service recommendation, which is to adeptly steer users towards services tailored to their distinct requirements and preferences. However, achieving accurate and real-time QoS predictions in the presence of various issues and anomalies, including outliers, data sparsity, grey sheep instances, and cold start scenarios, remains a challenge. Current state-of-the-art methods often fall short when addressing these issues simultaneously, resulting in performance degradation. In response, in this paper, we introduce anAnomaly-ResilientReal-timeQoSPrediction framework (called ARRQP). Our primary contributions encompass proposing an innovative approach to QoS prediction aimed at enhancing prediction accuracy, with a specific emphasis on improving resilience to anomalies in the data. ARRQP utilizes the power of graph convolution techniques, a powerful tool in graph-based machine learning, to capture intricate relationships and dependencies among users and services. By leveraging graph convolution, our framework enhances its ability to model and seize complex relationships within the data, even when the data is limited or sparse. ARRQP integrates both contextual information and collaborative insights, enabling a comprehensive understanding of user-service interactions. By utilizing robust loss functions, this approach effectively reduces the impact of outliers during the training of the predictive model. Additionally, we introduce a method for detecting grey sheep users or services that is resilient to sparsity. These grey sheep instances are subsequently treated separately for QoS prediction. Furthermore, we address the cold start problem as a distinct challenge by emphasizing contextual features over collaborative features. This approach allows us to effectively handle situations where newly introduced users or services lack historical data. Experimental results on the publicly available benchmark WS-DREAM 1 dataset demonstrate the framework's effectiveness in achieving accurate and timely QoS predictions, even in scenarios where anomalies abound.

The e-commerce industry has seen significant growth in recent years due to the introduction of new web service APIs. Quality-of-Service (QoS) parameters, which are fundamental for assessing service performance, have become crucial in evaluating services in the competitive market. Since QoS parameters can vary among users and change over time, accurate QoS predictions have become essential for users when selecting the most suitable services. Existing methods for predicting temporal QoS have hardly achieved the desired accuracy, beset by challenges like data sparsity, the presence of anomalies, and the inability to capture intricate temporal user-service interactions. Although some recent approaches, particularly those founded on recurrent neural network-based sequential architectures, endeavor to model temporal relationships in QoS data, they grapple with performance degradation due to the omission of other pivotal features, such as collaborative relationships and spatial characteristics of users and services. Furthermore, the uniform attention among features across all time-steps can thwart progress in predictive accuracy. This paper addresses these challenges and proffers a scalable strategy for temporal QoS prediction using multi-source collaborative features that not only furnishes heightened responsiveness but also engenders enhanced prediction accuracy. The method amalgamates collaborative features stemming from both users and services, capitalizing on the user-service relationship. Additionally, it integrates spatio-temporal auto-extracted features through the orchestration of graph convolution and a specialized variant of the transformer encoder equipped with multi-head self-attention. The proposed approach has been validated on the WSDREAM-2 benchmark datasets, and the results of these extensive experiments demonstrate that our framework surpasses major state-of-the-art methods in terms of predictive accuracy, all the while upholding robust scalability and reasonable responsiveness.

With the increasing trend of web services over the Internet, developing a robust Quality of Service (QoS) prediction algorithm for recommending services in real-time is becoming a challenge today. Designing an efficient QoS prediction algorithm achieving high accuracy, while supporting faster prediction to enable the algorithm to be integrated into a real-time system, is one of the primary focuses in the domain of Services Computing. The major state-of-the-art QoS prediction methods are yet to efficiently meet both criteria simultaneously, possibly due to the lack of analysis of challenges involved in designing the prediction algorithm. In this paper, we systematically analyze the various challenges associated with the QoS prediction algorithm and propose solution strategies to overcome the challenges, and thereby propose a novel offline framework using deep neural architectures for QoS prediction to achieve our goals. Our framework, on the one hand, handles the sparsity of the dataset, captures the non-linear relationship among data, figures out the correlation between users and services to achieve desirable prediction accuracy. On the other hand, our framework being an offline prediction strategy enables faster responsiveness. We performed extensive experiments on the publicly available WS-DREAM dataset to show the trade-off between prediction performance and prediction time. Furthermore, we observed our framework significantly improved one of the parameters (prediction accuracy or responsiveness) without considerably compromising the other as compared to the state-of-the-art methods.

With the proliferation of Internet-of-Things and continuous growth in the number of web services at the Internet-scale, the service recommendation is becoming a challenge nowadays. One of the prime aspects influencing the service recommendation is the Quality-of-Service (QoS) parameter, which depicts the performance of a web service. In general, the service provider furnishes the value of the QoS parameters before service deployment. However, in reality, the QoS values of service vary across different users, time, locations, etc. Therefore, estimating the QoS value of service before its execution is an important task, and thus, the QoS prediction has gained significant research attention. Multiple approaches are available in the literature for predicting service QoS. However, these approaches are yet to reach the desired accuracy level. In this article, we study the QoS prediction problem across different users, and propose a novel solution by taking into account the contextual (more specifically, location) information of both services and users. Our proposal includes two key steps: (a) hybrid filtering, and (b) hierarchical prediction mechanism. On the one hand, the hybrid filtering aims to obtain a set of similar users and services, given a target user and a service. On the other hand, the goal of the hierarchical prediction mechanism is to estimate the QoS value accurately by leveraging hierarchical neural-regression. We evaluated our framework on the publicly available WS-DREAM datasets. The experimental results show the outperformance of our framework over the major state-of-the-art approaches.

Efficient service composition in real time, while satisfying desirable Quality of Service (QoS) guarantees for the composite solution has been one of the topmost research challenges in the domain of services computing. On one hand, optimal QoS aware service composition algorithms, that come with the promise of solution optimality, are inherently compute intensive, and therefore, often fail to generate the optimal solution in real time for large scale web services. On the other hand, heuristic solutions that have the ability to generate solutions fast and handle large and complex service spaces, settle for sub-optimal solution quality. The problem of balancing the trade-off between computation efficiency and optimality in service composition has alluded researchers since quite some time, and several proposals for taming the scale and complexity of web service composition have been proposed in literature. In this paper, we present a new perspective towards this trade-off in service composition based on abstraction refinement, which can be seamlessly integrated on top of any off-the-shelf service composition method to tackle the space complexity, thereby, making it more time and space efficient. Instead of considering services individually during composition, we propose a set of abstractions and corresponding refinements to form service groups based on functional characteristics. The composition and QoS satisfying solution construction steps are carried out in the abstract service space. Our abstraction refinement methods give a significant speed-up compared to traditional composition techniques, since we end up exploring a substantially smaller space on average. Experimental results on benchmarks show the efficiency of our proposed mechanism in terms of time and the number of services considered for building the QoS satisfying composite solution.

Automatic web service composition has received a significant research attention in service-oriented computing over decades of research. With increasing number of web services, providing an end-to-end Quality of Service (QoS) guarantee in responding to user queries is becoming an important concern. Multiple QoS parameters (e.g., response time, latency, throughput, reliability, availability, success rate) are associated with a service, thereby, service composition with a large number of candidate services is a challenging multi-objective optimization problem. In this article, we study the multi-constrained multi-objective QoS-aware web service composition problem and propose three different approaches to solve the same, one optimal, based on Pareto front construction, and two others based on heuristically traversing the solution space. We compare the performance of the heuristics against the optimal and show the effectiveness of our proposals over other classical approaches for the same problem setting, with experiments on WSC-2009 and ICEBE-2005 datasets.

In this paper, we examine the problem of data dissemination and optimization in the context of a large scale distributed cyber-physical system (CPS), and propose a novel rule-based mechanism for effective observation collection and transmission. Our work rests on the idea that all observations on all parameters are not required at all times, and thereby, selective data transmission can reduce sensor workload significantly. Experiments show the efficacy of our proposal.

In recent times, automated business processes and web services have become ubiquitous in diverse application spaces. Efficient composition of web services in real time while providing necessary Quality of Service (QoS) guarantees is a computationally complex problem and several heuristic based approaches have been proposed to compose the services optimally. In this article, we present the design of a scalable QoS-aware service composition mechanism that balances the computational complexity of service composition with the QoS guarantees of the composed service and achieves scalability. Our design guarantees a single QoS parameter using an intelligent search and pruning mechanism in the composed service space. We also show that our methodology yields near optimal solutions on real benchmarks. We then enhance our proposed mechanism to guarantee multiple QoS parameters using aggregation techniques. Finally, we explore search time versus solution quality tradeoff using parameterized search algorithms that produce better-quality solutions at the cost of delay. We present experimental results to show the efficiency of our proposed mechanism.

In recent times, automated business processes and web service technologies have become popular and ubiquitous for catering to diverse user needs. While providing a service, the service providers are typically expected to furnish promised QoS values for the services they deliver. However, when the services are physically deployed and invoked during a query resolution, these parameter values vary largely depending on different factors like network load, number of applications running in the server etc. In this work, we present a stochastic model of the web service composition problem. Experimental results on Web Service Challenge (WSC) benchmarks show the efficiency of our proposed mechanism.

In this work, we propose a framework for construction of an approximate environment for compositional verification using invariants learned from dynamic traces of the system and the counterexamples generated by a model checker on verifying a property on the component in isolation. We adopt a counterexample ranking methodology for eliminating possibly fictitious counterexamples by choosing a minimal subset of the invariants. We explore the aspect of choosing a threshold for counterexamples as well as assume properties which can contribute towards further refining the subset chosen and produce a stronger abstraction. Experimental results on benchmark designs shows the efficacy of our proposal.

In recent times, automated business processes and web services have become ubiquitous in diverse application spaces. Efficient composition of web services in real time while providing necessary QoS guarantees is a computationally complex problem and several heuristic based approaches have been proposed to compose services optimally. In this paper, we present the design of a scalable but approximate QoS-aware service composition mechanism which balances the computational complexity of service composition with the QoS guarantees of the composed service and achieves scalability for dynamic service composition. We present experimental results to show the efficiency of our proposed mechanism.