Abstract The human brain, a critical organ within the central nervous system, is vulnerable to a range of complex and life-threatening disorders, including brain tumors, Alzheimer’s disease, and stroke. Accurate and timely diagnosis of these conditions is essential for effective treatment and management. Traditionally, brain disease detection relies on manual interpretation of medical imaging modalities such as magnetic resonance imaging (MRI), a process that is time-intensive, prone to human error, and often lacks consistency. To address these limitations, this study proposes an automated deep learning-based framework for brain disease classification using the concept of transfer learning. A comparative analysis of four advanced convolutional neural network (CNN) architectures, VGG-16, VGG-19, EfficientNet, and DenseNet121 was conducted to evaluate their diagnostic performance on a publicly available MRI dataset. To enhance generalization and prevent overfitting, data augmentation techniques were applied during the training phase. The proposed pipeline comprised data acquisition, preprocessing, and comprehensive model evaluation stratified into various training and testing splits. Performance was rigorously assessed using metrics including accuracy, precision, recall, specificity, and F1-score. The results demonstrate that the VGG-16-based approach surpassed the other state-of-the-art models in classification performance, showcasing its potential as a reliable tool for automated brain disease diagnosis. This work underscores the applicability of deep learning in neuroimaging analysis and opens avenues for future improvements with more advanced architectures and multimodal data integration.

Abstract Crowdsource software development has become more and more popular in recent years in the software industry. Crowdsourcing is an open‐call technique for outsourcing tasks to a broad and undefined crowd. Crowdsourcing provides numerous advantages including reduced costs, fast project completion, talent identification, diversity of solutions, top‐quality, and access to problem‐solving creativity. Despite of the benefits gained from crowdsourcing, there are numerous issues like lack of experienced workers, lack of confidentiality, copyright issues, software sustainability, and so forth. There is also less focus on the long‐term sustainability of software development because of new ideas emerging in crowdsourcing software development. Furthermore, in literature, lack of guidelines towards sustainable software crowdsourcing is highlighted as one of the limitations in the software standards. This study aims to identify the factors that influence sustainability aspects in crowdsourced software development. We have conducted a systematic literature review for identification of these factors. In this paper, we present findings of the systematic literature review in the form of a list of 11 factors extracted from a sample of 45 finally selected papers. Among these factors, six of the factors are ranked as critical factors. These critical factors are “Lack of coding standard in documentation,” “Use of popular programming tools,” “Crowd Lack of knowledge and awareness about sustainability,” “Energy‐efficient coding,” “Lack of awareness about sustainable software engineering practices,” and “Lack of coordination/communication between client and crowd.”

Abstract Requirement's elicitation is the process of gathering requirements from users, customers, and stakeholders using traditional or collaborative elicitation techniques. Software requirement gathering is a challenging task particularly in a Global Software Development (GSD) paradigm due to geographical distance, limited face to face meetings, time zone differences, and language and cultural barriers. As more companies begin to adopt GSD in order to save costs, the success factors need to be identified and evaluated in order to ensure a successful elicitation process. This paper offers an in‐depth analysis on success factors for requirement elicitation within GSD environment. First, all possible success factors are identified from the literature via a Systematic Literature Review (SLR). Next, these factors are evaluated by software industries via a questionnaire survey across different types and levels of experts, size of organizations, and from the client–vendor perspective. The relationships between the success factors and the survey results were evaluated using the Spearman's correlation coefficient. The results produced a 0.835 Spearman's correlation coefficient at significance level ρ = 0.000, which showed a strong positive correlation between the outcome of SLR and survey with no significant difference.

Information extraction and its service is a hot topic. Many works focus on extracting information from a certain web page and ignore the localization of the webpage which contains useful information. Nevertheless, developing a holistic system to extract information consists of locating a webpage and extracting information from that webpage, and these two steps are indispensable. For instance, extracting lecture news from universities' websites is a typical hard task that need to locate web pages and extract news information from them. Due to different layouts and visual appearances, statistic-based methods and visual based methods failed to find them. In this study, we propose an all-holistic method to locate lecture news on the university website. Graph Convolutional Network (GCN) is applied to fuse the multimodal data, which could learn useful features from different views, the linked relationship, the visual similarity, and the semantic of web pages. Firstly, we apply the link model to explore the parent-child relationship between web pages, then calculate the similarity of parent-child pages using a visual model and obtain the semantic features based on the BERT model. Specifically, the visual similarity features are learned based on triplet loss function which imposes the Convolutional Neural Network (CNN) model to learn similar parts in the same group. Lastly, these features are fused into the GCN model to find a certain webpage and it can be adaptive to various university websites. The experiments conducted on 50 websites show our method outperforms state-of-the-art.

With the rapid growth of internet usage for enterprise-wide and cross-enterprise business applications (such as those in Electronic Commerce), workflow systems are gaining importance as an infrastructure for automating inter-organizational interactions. However, the traditional centralized workflow management technology can no longer meet the needs of current application services. For example, in e-commerce, cross-enterprise business applications may cause many security problems and cross-domain problems in the implementation of workflow. At the same time, due to the uncertainty and variability of environment and user requirements in practical applications, many business logics are difficult to be completely defined in advance. Therefore workflow models need to be immediately built or adjusted dynamically. Nowadays, distributed scheduling and decentralized control of workflow have become the emerging trend and are facing many challenges at the forefront of Internet development technology. In this paper, a distributed workflow control execution method based on “meta-process” is proposed. Specifically, we designed and implemented a decentralized distributed scheduling management system for workflow tasks. To manage and control the distributed scheduling of workflow, we constructed a “meta-process”, which can ensure the integrity of the control chain in the distributed scheduling process. Our system can efficiently handle the data state migration between task nodes and supports the dynamic adjustment of the workflow model. For validation, we simulated a large number of service scheme samples and applied them to the system, which proved that all service cases can be executed correctly. Therefore, the feasibility of this method is verified.

Information service is always a hot topic especially when the Web is accessible anywhere. In university, lecture information is very important for students and teachers who want to take part in academic meetings. Therefore, lecture news extraction is an important and imperative task. Many open information extraction methods have been proposed, but due to the high heterogeneity of websites, this task is still a challenge. In this paper, we propose a method based on fusing multiple features to locate lecture news on the university website. These features include the linked relationship between parent webpage and child webpages, the visual similarity, and the semantics of webpages. Additionally, this paper provides an information service based on a main content extraction algorithm for extracting the lecture information. Stable and invariant features enable the proposed method to adapt to various kinds of campus websites. The experiments conducted on 50 websites show the effectiveness and efficiency of the provided service.

Nowadays, an Internet of Things (IoT) device consists of algorithms, datasets, and models. Due to good performance of deep learning methods, many devices integrated well-trained models in them. IoT empowers users to communicate and control physical devices to achieve vital information. However, these models are vulnerable to adversarial attacks, which largely bring potential risks to the normal application of deep learning methods. For instance, very little changes even one point in the IoT time-series data could lead to unreliable or wrong decisions. Moreover, these changes could be deliberately generated by following an adversarial attack strategy. We propose a robust IoT data classification model based on an encode-decode joint training model. Furthermore, thermometer encoding is taken as a nonlinear transformation to the original training examples that are used to reconstruct original time series examples through the encode-decode model. The trained ResNet model based on reconstruction examples is more robust to the adversarial attack. Experiments show that the trained model can successfully resist to fast gradient sign method attack to some extent and improve the security of the time series data classification model.

Anomaly detection algorithms (ADA) have been widely used as services in many maintenance monitoring platforms. However, there are numerous algorithms that could be applied to these fast changing stream data. Furthermore, in IoT stream data due to its dynamic nature, the phenomena of conception drift happened. Therefore, it is a challenging task to choose a suitable anomaly detection service (ADS) in real time. For accurate online anomalous data detection, this paper developed a service selection method to select and configure ADS at run-time. Initially, a time-series feature extractor (Tsfresh) and a genetic algorithm-based feature selection method are applied to swiftly extract dominant features which act as representation for the stream data patterns. Additionally, stream data and various efficient algorithms are collected as our historical data. A fast classification model based on XGBoost is trained to record stream data features to detect appropriate ADS dynamically at run-time. These methods help to choose suitable service and their respective configuration based on the patterns of stream data. The features used to describe and reflect time-series data’s intrinsic characteristics are the main success factor in our framework. Consequently, experiments are conducted to evaluate the effectiveness of features closed by genetic algorithm. Experimentations on both artificial and real datasets demonstrate that the accuracy of our proposed method outperforms various advanced approaches and can choose appropriate service in different scenarios efficiently.

Abstract Usage of agile methods for software development has increased in recent times. Rapid delivery of software products is ensured by these methods with less expense and high user gratification. Application of these methods on a large scale causes many De‐Motivators or challenges as initially these methods were aimed at those developmental teams that worked on small scale. However, no effort has been made yet to hand out the various correlations among the challenges. To fill the space, the Interpretive Structural Model (ISM) approach is used to reconnoiter the correlations among the De‐Motivators regarding Scaling Agile Software Development Methodologies (ASDM) from the management perspective. Creation of a framework for structural modeling association among the De‐Motivators is the aim of this study. To attain the goal, a hybrid methodology was applied based on a Systematic literature review (SLR), empirical survey, and ISM. First, we pointed out 15 De‐Motivators of Scaling ASDM from management perspectives through SLR study. Second, with the help of 59 experts from different countries, a questionnaire survey was conducted to empirically examine the correlations among the detected De‐Motivators. Further, correlations among the challenges are discovered with the usage of ISM through review of the panel, and they accomplish the categorization through CrossImpact Matrix Multiplication Applied Approach.

Digital divide means unequal access to the people for information and communication technology (ICT) facilities. The developed countries are comparatively less digitally divided as compared to developing countries. This study focuses on District Chitral considering its geographical conditions and high mountainous topography which plays a significant role in its isolation. Aside from the digital divide, the situation in Chitral is even more severe in terms of the absence of basic ICT infrastructure and electricity in the schools. To address this issue, especially in female secondary and higher secondary schools, we designed a project to bridge the digital divide via Wireless Local Area Network on Raspberry Pi3 for balancing the ICT facilities in the targeted area. The Wi‐Fi‐Based Content Distributors (Wi‐Fi‐BCDs) were provided to bridge the digital divide in rural area schools of Chitral. The Wi‐Fi‐BCD is a solar‐based system that is used to deliver quality educational contents directly to classroom, library, or other learning environments without electricity connection and Internet wire as these facilities are available by default in it. The close‐ended questionnaire was adopted to collect data from the students, teachers, and headmistresses of girl secondary and higher secondary schools in Chitral. The procedure of validity, reliability, regression, correlation, and exploratory factor analysis was used to analyze the obtained data. The technology acceptance model (TAM) was modified and adopted to examine the effects of Wi‐Fi‐BCD for bridging the digital divide. The relationship of the modified TAM model was examined through regression and correlation to verify the model fitness according to the data obtained. The result analysis of this study shows that the relationship of the modified TAM model with its variables is positively significant, while the analysis of path relationship between model variables and outcomes from the questionnaire shows that it motivates learners to use Wi‐Fi‐BCD.

A challenged network is one where traditional hypotheses such as reduced data transfer error rates, end‐to‐end connectivity, or short transmissions have not gained much significance. A wide range of application scenarios are associated with such networks. Delay tolerant networking (DTN) is an approach that pursues to report the problems which reduce communication in disrupted networks. DTN works on store‐carry and forward mechanism in such a way that a message may be stored by a node for a comparatively large amount of time and carry it until a proper forwarding opportunity appears. To store a message for long delays, a proper buffer management scheme is required to select a message for dropping upon buffer overflow. Every time dropping messages lead towards the wastage of valuable resources which the message has already consumed. The proposed solution is a size‐based policy which determines an inception size for the selection of message for deletion as buffer becomes overflow. The basic theme behind this scheme is that by determining the exact buffer space requirement, one can easily select a message of an appropriate size to be discarded. By doing so, it can overcome unnecessary message drop and ignores biasness just before selection of specific sized message. The proposed scheme Spontaneous Size Drop (SS‐Drop) implies a simple but intelligent mechanism to determine the inception size to drop a message upon overflow of the buffer. After simulation in ONE (Opportunistic Network Environment) simulator, the SS‐Drop outperforms the opponent drop policies in terms of high delivery ratio by giving 66.3% delivery probability value and minimizes the overhead ratio up to 41.25%. SS‐Drop also showed a prominent reduction in dropping of messages and buffer time average.

Abstract Most of the software development projects have traditionally been faced with severe problems in terms of quality, cost, and time. Researchers and practitioners have focused on agile software development methods (ASDMs) as an alternative to overcome these problems. Agile methods employ iterative development cycles, interspersed by user feedback. Agile methods were basically developed for small development teams. Scaling agile methods is a big issue from different perspectives. De‐motivators play a key role in project management as it allows early identification and prompt management of threats that may arise during project execution. The objective of this paper is to identify the de‐motivators while scaling agile at large, from management perspectives. We have adapted SLR and applied contrived search criteria derived from the research questions, followed by selecting the required research papers, data extraction, and data synthesis, which resulted in 15 de‐motivators from 58 relevant papers. Some of the identified de‐motivators are ‘traditional organizational culture’, ‘lack of agile experts’, ‘reluctance to adopt’, and ‘lack of management and commitment support’. The identified factors have been compared from various perspectives, such as continents, digital libraries, organization size, and so forth.

Abstract Software Outsourcing Partnership (SOP) is considered as a type of risk and reward sharing relationship between a client organisation, in the developed countries, and its overseas vendor organisation. Regardless of numerous benefits, the development of SOP still remnants in its infancy stage due to several interactive barriers. Some studies have been conducted to examine the barriers to SOP formation. However, no attempt has been reported so far to explore the multifaceted interrelationships amongst them. To bridge the gap, this study implements Interpretive Structural Model (ISM) approach to reconnoitre the interrelationships amongst the barriers in the context of SOP formation. The objective of this research paper is to develop a framework for modelling structural association amongst the barriers. To achieve the objective, we used a hybrid methodology based on systematic literature review (SLR), empirical survey, and ISM. Firstly, via SLR study, we identified 27 barriers to SOP formation. Secondly, to empirically explore the interrelationships amongst the identified barriers, a questionnaire survey was performed with 50 experts from a total of 20 different countries. Further, interrelationships amongst the barriers were identified using ISM via panel review, and their classifications were carried out via Cross‐Impact Matrix Multiplication Applied to the Classification Approach.

Abstract Software Outsourcing Partnership (SOP) is considered as a kind of risk and reward sharing client‐vendor relationship. Generally, a fruitful outsourcing association might be converted to an outsourcing partnership. The objective of this research is to identify and analyse barriers that are hurdles to vendors in renewing or promoting their ongoing client‐vendor relationship to outsourcing partnership. A questionnaire survey based on the findings of Systematic Literature Review (SLR) was performed with 50 experts. The study identifies five critical barriers such as “insufficient quality of technical capability,” “poor infrastructure,” “poor quality of service,” “communication gap and poor coordination,” and “relational risk.” The results indicate that barriers' insufficient quality of technical capability, poor infrastructure, and poor quality of service were common in four types of experts while insufficient quality of technical capability is common in three levels of experts. Furthermore, barriers were classified based on their criticality from client‐vendor perspective. The results of Spearman correlation test (rs = 0.714 and ρ = 0.000) confirmed that the participant strongly agrees with the outcomes of the SLR. The results suggest that for successful renewal or promotion of their existing outsourcing association, vendor organizations should address all the identified barriers in general and the most common barriers in particular.

Borehole lithology discrimination is the foundation for formation evaluation and reservoir characterization. Due to the limitation of costing or accuracy, direct discrimination methods, such as borehole core and drilling cutting analysis, are unable to widely apply, while logging lithology interpretation provides an alternative solution for this task. To mitigate the influence of subjective bias, several machine learning algorithms, such as neural network, support vector machine, decision tree, and random forest (RF), have already been applied for logging lithology interpretation. However, the vast majority of preceding studies are simple applications that directly apply classification algorithms to the raw input space formed by logging curve values, only limited studies involved feature extraction or learning space transformation. In this letter, we propose a hybrid algorithm that combines the mean-shift algorithm and the RF algorithm for borehole lithology discrimination in the prototype similarity space. Experiments on data collected from nine different areas demonstrate that the proposed algorithm has significant advantages in accuracy compared with other algorithms, which provides a considerable alternative way for further machine learning-assisted logging lithology interpretation.

Software outsourcing partnership (SOP) is mutually trusted inter-organisational software development relationship between client and vendor organisations based on shared risks and benefits. SOP is different to conventional software development outsourcing relationship, SOP could be considered as a long term relation with mutual adjustment and renegotiations of tasks and commitment that exceed mere contractual obligations stated in an initial phase of the collaboration. The objective of this research is to identify various factors that are significant for vendors in conversion of their existing outsourcing contractual relationship to partnership. We have performed a systematic literature review for identification of the factors. We have identified a list of factors such as 'mutual interdependence and shared values', 'mutual trust', 'effective and timely communication', 'organisational proximity' and 'quality production' that play vital role in conversion of the existing outsourcing relationship to a partnership.