Abstract

Walking in a Virtual Environment is a bounded task. It is challenging for a subject to navigate a large virtual environment designed in a limited physical space. External hardware support may be required to achieve such an act in a concise physical area without compromising navigation and virtual scene rendering quality. This paper proposes an algorithmic approach to let a subject navigate a limitless virtual environment within a limited physical space with no additional external hardware support apart from the the regular Head-Mounted-Device (HMD) itself. As part of our work, we developed a Virtual Art Gallery as a use-case to validate our algorithm. We conducted a simple user-study to gather feedback from the participants to evaluate the ease of locomotion of the application. The results showed that our algorithm could generate limitless paths of our use-case under predefined conditions and can be extended to other use-cases

Abstract

The adoption of Machine Learning (ML) in software applications has increased in domains like healthcare, banking and others. leading to coining of the term MLWare applications. However, challenges like diverse code base, complex components, lack of expertise, etc. persist in development and maintenance of these applications. Application of software engineering patterns and practices for the development of MLWare applications can improve maintainability, extensibility, scalability and other software quality parameters. In this paper, we propose an approach for developing MLWare applications using a pattern oriented approach. We demonstrate the approach on a credit risk scorecard application that can helps loan officer identify risk patterns and make loan decisions.

Abstract

Requirements Engineering in the industry is expertise-driven, heavily manual, and centered around various types of requirement specification documents being prepared and maintained. These specification documents are in diverse formats and vary depending on whether it is a business requirement document, functional specification, interface specification, client specification, and so on. These diverse specification documents embed crucial product knowledge such as functional decomposition of the domain into features, feature hierarchy, feature types and their specific feature characteristics, dependencies, business context, etc. Moreover, in a product development scenario, thousands of pages of requirement specification documentation is created over the years. Comprehending functionality and its associated context from large volumes of specification documents is a highly complex task. To address this problem, we propose to digitalize the requirement specification documents into processable models. This paper discusses the salient aspects involved in the digitalization of requirements knowledge from diverse requirement specification documents. It proposes an AI engine for the automatic transformation of diverse text-based requirement specifications into machine-processable models using NLP techniques and the generation of context-sensitive user stories. The paper describes the key requirement abstractions and concepts essential in an industrial scenario, the conceptual meta-model, and DizReq engine (AI engine for digitalizing requirements) implementation for automatically transforming diverse requirement specifications into user stories embedding the business context. The evaluation results from digitalizing specifications of an IT product suite are discussed: mean feature extraction efficiency is 40 features/file, mean user story extraction efficiency is 71 user stories/file, feature extraction accuracy is 94%, and requirement extraction accuracy is 98%. Keywords: NLP, NLP4RE, model extraction, requirements digitalization, requireme

Abstract

With increasing computing power, capacity and internet connectivity, the ability to learn has crossed classroom boundaries in the form of online learning. The adoption rate and effectiveness of online learning continues to be an area of research. Collaborative and Cooperative (CoCo) learning involves a joint effort by students to understand and solve problems or tasks. Adoption of open source software has also increased with technology-aided collaboration and cooperation. Leveraging the practices of crowdsourced software development, we propose CoCo Learning approach for technical education using virtual labs. About 145 undergraduate engineering students from various institutes participated in our study for a period of one full semester. About 92% of the students conducted an average of 24 virtual experiments with this approach. Nearly 90% of the faculty from these institutes also confirmed that their students' understanding of courses has improved while delivering tangible software artifacts as outcomes.

Abstract

Quality Models play a critical role in assuring quality and have evolved over 40+ years. They provide support for defining quality attributes, building and measuring the quality of the resulting product. Each quality model adopts a critical view on quality in terms of a set of model elements and relationships between them. This study aims to provide an overview of the state-of-the-art research on quality models with a focus on encompassing model elements and their support to architecting quality. The study was conducted using systematic mapping as the research methodology. A total of 238 primary papers were classified based on the type of research, standards usage, and publication trends. We identified that 17% (40) of papers belong to quality models. These 40 models were analyzed for the underlying meta-model elements and their support for a quality architecture using   rence architecture framework. The architecture phase mapping analysis shows that quality planning phase is 100% supported, quality assessment is 75% supported, quality documentation is included in 40% models and quality realization aspect is barely considered in 13% models. Quality realization happens through software processes and patterns, and it is necessary to evolve quality models and software process architectures that correlate quality definitions and quality realization mechanisms. Future research is expected in this direction

Abstract

With the rise of Virtual Reality (VR) footprint in many organizations, it was unclear if traditional software engineering practices are still exercised during VR product development. As part of our research, we conducted a year-long multi-level exploratory study to understand the various software development practices within VR product development teams. An empirical study on VR practitioners from 6 different countries was done to examine their development strategies, methods, and models adopted along with the various challenges faced during the course of VR product release. We found that VR practitioners adopted hybrid Software engineering approaches in VR product development. In this paper, we present our insights from the empirical study and stress on the need for a diverse software development model for VR products.

Abstract

VR development practices have a diverse set of practices compared to traditional software development. Tasks like scene design, acoustic design, vergence manipulation, image depth, etc. are specific to VR apps and hence require evaluation processes that may be different from the traditional means. Usability Evaluation is one such process which is being executed in an unconventional way by Industrial Practitioners today. In this paper, the researchers detail a Systematic Literature Review of the Usability Evaluation Methods practised by Industrial researchers while building VR Products. The researchers found that VR Product teams follow unique methods to improve usability in their products. Further, the researchers consolidate these methods and provide insights into choosing the best to build a real-world VR Product based on the defined product constraints.

Abstract

Identifying dependency between various artifacts in a large scale software system is a non-trivial task. As the software evolves, multiple artifacts like files, docs, classes, database scripts, etc., are likely to undergo change concurrently. Such artifacts tend to have a dependency between them, otherwise referred to as logical coupling. Researchers have used Support and Confidence as an association rule based measurement to predict the levels of logical coupling among the software artifacts. However, employing a single change on a software artifact can span across various closely related changes when many code contributors are working on the same change. Thus it is important to preprocess and group these semantically related change-sets before identifying logical coupling. In this paper, we propose a method to identify logical coupling and group semantically related change sets. We evaluate our method on real-world git repositories and document our observations

Abstract

With more than a trillion web pages, there is a plethora of content available for consumption. Search Engine queries invariably lead to overwhelming information, parts of it relevant and some others irrelevant. Often the information provided can be conflicting, ambiguous, and inconsistent contributing to the loss of credibility of the content. In the past, researchers have proposed approaches for credibility assessment and enumerated factors influencing the credibility of web pages. In this work, we detailed a W EBCred framework for automated genre-aware credibility assessment of web pages. We developed a tool based on the proposed framework to extract web page features instances and identify genre a web page belongs to while assessing it’s Genre Credibility Score (GCS). We validated our approach on ‘Information Security’ dataset of 8,550 URLs with 171 features across 7 genres. The supervised learning algorithm, Gradient Boosted Decision Tree classified genres with 88.75% testing accuracy over 10 fold cross-validation, an improvement over the current benchmark. We also examined our approach on ‘Health’ domain web pages and had comparable results. The calculated GCS correlated 69% with crowdsourced Web Of Trust (W OT) score and 13% with algorithm based Alexa ranking across 5 Information security groups. This variance in correlation states that our GCS approach aligns with human way (W OT) as compared to algorithmic way (Alexa) of web assessment in both the experiments.

Abstract

Software Engineering Researchers in India from both academia and industry are widely contributing to various research problems. In this report, we briefly summarize the key insights from 1st Software Engineering Research in India Update Meeting (SERI 2019), which provides a platform for all the researchers to present and discuss their research work. The essence of this research meeting is to examine the current\ research areas within the software engineering research community and discover the scope of collaboration. Specifically, the first SERI update meeting had a series of invited research talks presented by researchers from academia and industry representing various software engineering labs across India. These talks unveiled notable views through presentations and group discussions, and cogitation of topics that can lead to fruitful collaborations between software engineering researchers and industry practitioners. The main objective of this update meeting is the exchange of research areas that promotes successful short-term and long-term collaborations.