Sanjay Podder

Towards Accurate Duplicate Bug Retrieval Using Deep Learning Techniques

Duplicate Bug Detection is the problem of identifying whether a newly reported bug is a duplicate of an existing bug in the system and retrieving the original or similar bugs from the past. This is required to avoid costly rediscovery and redundant work. In typical software projects, the number of duplicate bugs reported may run into the order of thousands, making it expensive in terms of cost and time for manual intervention. This makes the problem of duplicate or similar bug detection an important one in Software Engineering domain. However, an automated solution for the same is not quite accurate yet in practice, in spite of many reported approaches using various machine learning techniques. In this work, we propose a retrieval and classification model using Siamese Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) for accurate detection and retrieval of duplicate and similar bugs. We report an accuracy close to 90% and recall rate close to 80%, which makes possible the practical use of such a system. We describe our model in detail along with related discussions from the Deep Learning domain. By presenting the detailed experimental results, we illustrate the effectiveness of the model in practical systems, including for repositories for which supervised training data is not available.

Topic cohesion preserving requirements clustering

This paper focuses on the problem of generating human interpretable clusters of semantically related plain-text requirements. Presented approach applies techniques from information retrieval, natural language processing, network analysis, and machine learning for identifying semantically central terms as themes and clustering requirements into semantically coherent groups together with meaningful explanatory themes associated with the clusters to assist in user comprehension of the clusters. Presented approach is generic in nature and can be used for other phases of SDLC (Software Development Life Cycle) including code-comprehension and architectural discovery. Suggested approach is particularly suitable for developing automated tool support for requirements management and analysis.

Domain Ontology Induction Using Word Embeddings

Ontology, the shared formal conceptualization of domain information, has been shown to have multiple applications in modeling, processing and understanding natural language text. In this work, we use distributed word vectors out of various recent language models from Deep Learning for semi-automated domain ontology creation for closed domains. We cover all major aspects of Domain Ontology Induction or Learning like concept identification, attribute identification, taxonomical and non-taxonomical relationship identification using the distributed word vectors. Preliminary results show that simple clustering based methods using distributed word vectors from these language models outperforms methods using models like LSI in ontology learning for closed domains.

Identifying implementation bugs in machine learning based image classifiers using metamorphic testing

We have recently witnessed tremendous success of Machine Learning (ML) in practical applications. Computer vision, speech recognition and language translation have all seen a near human level performance. We expect, in the near future, most business applications will have some form of ML. However, testing such applications is extremely challenging and would be very expensive if we follow todays methodologies. In this work, we present an articulation of the challenges in testing ML based applications. We then present our solution approach, based on the concept of Metamorphic Testing, which aims to identify implementation bugs in ML based image classifiers. We have developed metamorphic relations for an application based on Support Vector Machine and a Deep Learning based application. Empirical validation showed that our approach was able to catch 71% of the implementation bugs in the ML applications.

An immersive future for software engineering: avenues and approaches

Software systems are increasingly becoming more intricate and complex, necessitating new ways to be able to comprehend and visualize them. At the same time, the nature of software engineering teams itself is changing with people playing more fluid roles often needing seamless and contextual intelligence, for faster and better decisions. Moreover, the next-generation of software engineers will all be post-millennials, which may have totally different expectations from their software engineering workplace. Thus, we believe that it is important to have a re-look at the way we traditionally do software engineering and immersive technologies have a huge potential here to help out with such challenges. However, while immersive technologies, devices and platforms, have matured in past few years, there has been very little research on studying how these technologies can influence software engineering. In this paper, we introduce how traditional software engineering can leverage immersive approaches for building, delivering and maintaining next-generation software applications. As part of our initial research, we present an augmented-reality based prototype for project managers, which provides contextual and immersive insights. Finally, we also discuss important research questions that we are investigating further as part of our immersive software engineering research.

Compliance adherence in distributed software delivery: a blockchain approach

In this extended abstract, we propose a conceptual framework that leverages distributed ledger technology and smart contracts to create a decentralized system to capture the occurrence of interesting development activities (e.g., a development build) and associated contextual data, and automatically audit and evaluate compliance to governance policies. Our hypothesis is that such a framework will facilitate easier sharing of information across all participants of a distributed development team, compliance evaluation and early mitigation actions, leading to greater visibility and compliance. Currently, the proof of concept we are working on is focused on sharing and compliance evaluation of the open-source components used in software development.

Data-Driven Application Maintenance: Views from the Trenches.

In this paper we present our experience during design, development, and pilot deployments of a data-driven machine learning based application maintenance solution. We implemented a proof of concept to address a spectrum of interrelated problems encountered in application maintenance projects including duplicate incident ticket identification, assignee recommendation, theme mining, and mapping of incidents to business processes. In the context of IT services, these problems are frequently encountered, yet there is a gap in bringing automation and optimization. Despite long-standing research around mining and analysis of software repositories, such research outputs are not adopted well in practice due to the constraints these solutions impose on the users. We discuss need for designing pragmatic solutions with low barriers to adoption and addressing right level of complexity of problems with respect to underlying business constraints and nature of data.

Evaluating complexity and digitizability of regulations and contracts for a blockchain application design

Blockchain technology becomes the key solution to provide trust and security without any need for a central supervisory authority to validate the transactions. By now, it plays a key role in the digital transformation of several processes and industries with varying application use cases. To promote the wide adoption of blockchain technology we need mechanisms to identify the digitizability level of the given regulations to smart contracts and mechanisms to specify which blockchain technology is best suitable for the given regulations. In this work, we propose a modeling approach that supports the automated analysis of human-readable regulation representations by suggesting how much percentage of regulation is digitizable and the suitable blockchain environment to design the application. We identify smart contract components that correspond to real-world entities and its pertaining clauses and its digitizability property. With selected examples, we explore this capability and discuss our future research directions on smart contract generation according to the recommended environment.

Accelerating Test Automation through a Domain Specific Language

Test automation involves the automatic execution of test scripts instead of being manually run. This significantly reduces the amount of manual effort needed and thus is of great interest to the software testing industry. There are two key problems in the existing tools & methods for test automation - a) Creating an automation test script is essentially a code development task, which most testers are not trained on, and b) the automation test script is seldom readable, making the task of maintenance an effort intensive process. We present the Accelerating Test Automation Platform (ATAP) which is aimed at making test automation accessible to non-programmers. ATAP allows the creation of an automation test script through a domain specific language based on English. The English-like test scripts are automatically converted to machine executable code using Selenium WebDriver. ATAPs English-like test script makes it easy for non-programmers to author. The functional flow of an ATAP script is easy to understand as well thus making maintenance simpler (you can understand the flow of the test script when you revisit it many months later). ATAP has been built around the Eclipse ecosystem and has been used in a real-life testing project. We present the details of the implementation of ATAP and the results from its usage in practice.