Question answering is one of the primary challenges of natural language understanding. In realizing such a system, providing complex long answers to questions is a challenging task as opposed to factoid answering as the former needs context disambiguation. The different methods explored in the literature can be broadly classified into three categories namely: 1) classification based, 2) knowledge graph based and 3) retrieval based. Individually, none of them address the need of an enterprise wide assistance system for an IT support and maintenance domain. In this domain, the variance of answers is large ranging from factoid to structured operating procedures; the knowledge is present across heterogeneous data sources like application specific documentation, ticket management systems and any single technique for a general purpose assistance is unable to scale for such a landscape. To address this, we have built a cognitive platform with capabilities adopted for this domain. Further, we have built a general purpose question answering system leveraging the platform that can be instantiated for multiple products, technologies in the support domain. The system uses a novel hybrid answering model that orchestrates across a deep learning classifier, a knowledge graph based context disambiguation module and a sophisticated bag-of-words search system. This orchestration performs context switching for a provided question and also does a smooth hand-off of the question to a human expert if none of the automated techniques can provide a confident answer. This system has been deployed across 675 internal enterprise IT support and maintenance projects.

This research focuses on analyzing and improving maintenance process by exploring novel applications of process mining and predictive analytics. We analyze the software maintenance process by applying process mining on software repositories, and address the identified inefficiencies using predictive analytics. To drive our research, we engage with practitioners from large, global IT companies and emphasize on the practical usability of the proposed solution approaches, which are evaluated by conducting a series of case studies on open source and commercial projects.

Mathematical word problems (or story problems) allow students to apply their mathematical problem solving ability to other subjects and real-world situations. Word problems build higher-order thinking, critical problem-solving, and reasoning skills. Generally solving a word problem is associated with mathematical modeling of a real word situation or a concept of another subject which is embedded in the problem. Manually creating word problems require knowledge of other topics a student is learning in parallel. Besides this, modeling mathematics with some other dissociated concept is a time-consuming and labor-intensive task. Due to lack of this integrated knowledge of other topics being taught, the substantive breadth of word problems is often very narrow and is limited to very few concepts. To address this limitation, we built a tool called Intelligent Math Tutor (IMT), which automatically generates mathematical word problems such that teachings from other subjects from a given curriculum can also be incorporated. Our tool thus widens the scope of word problems and uses this problem-solving based approach to indirectly create cognizance in its students. To the best of our knowledge, our tool is the first of its kind tool which explicitly blends knowledge from multiple dissociated subjects and uses it to enhance the cognizance of its learners.

Process mining consists of mining event logs generated from business process execution supported by Information Systems (IS). Process mining of software repositories has diverse applications because vast data is generated during Software Development Life Cycle (SDLC) and archived in IS such as Version Control System (VCS), Peer Code Review (PCR) System, Issue Tracking System (ITS), and mail archives. There is need to explore its applications on different repositories to aid managers in process management. We conduct two phase surveys and interviews with managers in a large, global, IT company. The first survey and in-person interviews identify the process challenges encountered by them that can be addressed by novel applications of process mining. We filter, group and abstract responses formulating 30 generic problem statements. On the basis of process mining type, we classify identified problems to eight categories such as control analysis, organizational analysis, conformance analysis, and preventive analysis. The second survey asks distinct participants the importance of solving identified problems. We calculate proposed Net Importance Metric (NIM) using 1262 ratings from 43 participants. Combined analysis of NIM and first survey responses reveals that the problems mentioned by few practitioners in first survey are considered important by majority in the second survey. We elaborate on possible solutions and challenges for most frequent and important problems. We believe solving these validated problems will help managers in improving project quality and productivity.

Online Analytical Processing (OLAP) cube is a multi-dimensional dataset used for analyzing data in a Data Warehouse (DW) for the purpose of extracting actionable intelligence. Process mining consists of analyzing event log data produced from Process Aware Information Systems (PAIS) for the purpose of discovering and improving business processes. Process cube is a concept which falls at the intersection of OLAP cube and process mining. Process cube facilitates process mining from multiple-dimensions and enables comparison of process mining results across various dimensions. We present an application of process cube to software defect resolution process to analyze and compare process data from a multi-dimensional perspective. We present a framework, a novel perspective to mine software repositories using process cube. Each cell of process cube is defined by metrics from multiple process mining perspectives like control flow, time, conformance and organizational perspective. We conduct a case-study on Google Chromium project data in which the software defect resolution process spans three software repositories: Issue Tracking System (ITS), Peer Code Review System (PCR) and Version Control System (VCS). We define process cube with 9 dimensions as issue report timestamp, priority, state, closed status, OS, component, bug type, reporter and owner. We define hierarchies along various dimensions and cluster members to handle sparsity. We apply OLAP cube operations such as slice, dice, roll-up and drill-down, and create materialized sub log for each cell. We demonstrate the solution approach by discovering process map and compare process mining results from Control Flow and Time perspective for Performance and Security issues.

Issue reporting and resolution is a software engineering process supported by tools such as Issue Tracking System (ITS), Peer Code Review (PCR) system and Version Control System (VCS). Several open source software projects such as Google Chromium and Android follow process in which a defect or feature enhancement request is reported to an issue tracker followed by source-code change or patch review and patch commit using a version control system. We present an application of process mining three software repositories (ITS, PCR and VCS) from control flow and organizational perspective for effective process management. ITS, PCR and VCS are not explicitly linked so we implement regular expression based heuristics to integrate data from three repositories for Google Chromium project. We define activities such as bug reporting, bug fixing, bug verification, patch submission, patch review, and source code commit and create an event log of the bug resolution process. The extracted event log contains audit trail data such as caseID, timestamp, activity name and performer. We discover runtime process model for bug resolution process spanning three repositories using process mining tool, Disco, and conduct process performance and efficiency analysis. We identify bottlenecks, define and detect basic and composite anti-patterns. In addition to control flow analysis, we mine event log to perform organizational analysis and discover metrics such as handover of work, subcontracting, joint cases and joint activities.

Success of a Q&A forum depends on volume of content (questions and answers) and quality of content (are the questions asked relevant, answers provided correct etc). Community participation is essential to create and curate content. Since their inception in 2008, stack exchange based forums have been able to engage a large number of users to create a rich repository of good quality questions and answers. In this paper, we wish to investigate the “activeness” of users in the stackexchange network particularly from a perspective of content creation. We also attempt to measure how the forums' incentive mechanism has enabled user's activeness. Further, we investigate how user's have diffused to other parts of the stack exchange network over time, hence bootstrapping new forums.

An analysis of commit history of Android reveals that Android has a code base of 550K files, where on an average each file has been modified 8.7 times. 41% of files have been modified at-least once. In terms of contributors, it has an overall contributor community of 1563, with 58.5% of them having made >; 5 commits. Moreover, the contributor community shows high churn levels, with only 13 of contributors continuing from 2005 to 2011. In terms of industry participation, Google & Android account for 22% of developers. Intel and RedHat account for 2% of contributors each and IBM, Oracle, TI, SGI account for another 1% each. Android code can be classified into 5 sub-projects: kernel, platform, device, tools and toolchain. In this paper, we profile each of these sub-projects in terms of change volumes, contributor and industry participation. We further picked specific framework topics such as UI, security, whose understanding is required from perspective of developing apps over Android, and present some insights on community participation around the same.

Given the high costs of software testing and fixing bugs after release, early detection of bugs using static analysis can result in significant savings. However, despite their many benefits, recent availability of many such tools, and evidence of a positive return-on-investment, static-analysis tools are not used widely because of various usability and usefulness problems. The usability inhibitors include the lack of features, such as capabilities to merge reports from multiple tools and view warning deltas between two builds of a system. The usefulness problems are related primarily to the accuracy of the tools: identification of false positives (or, spurious bugs) and uninteresting bugs among the true positives. In this paper, we present the details of an online portal, developed at IBM Research, to address these problems and promote the adoption of static-analysis tools. We report our experience with the deployment of the portal within the IBM developer community. We also highlight the problems that we have learned are important to address, and present our approach toward solving some of those problems.