A5057662526- Software Engineering Research
- EEG and Brain-Computer Interfaces
- Gaze Tracking and Assistive Technology
- Topic Modeling
- Semantic Web and Ontologies
- Natural Language Processing Techniques
- Software Engineering Techniques and Practices
- Reading and Literacy Development
- Tactile and Sensory Interactions
- Software System Performance and Reliability
- Neural and Behavioral Psychology Studies
- Alzheimer's disease research and treatments
- Neurobiology of Language and Bilingualism
- Hearing Loss and Rehabilitation
- Functional Brain Connectivity Studies
- Advanced Software Engineering Methodologies
- Metabolomics and Mass Spectrometry Studies
- Bioinformatics and Genomic Networks
- Intelligent Tutoring Systems and Adaptive Learning
- Machine Learning in Materials Science
- Visual Attention and Saliency Detection
- Software Testing and Debugging Techniques
- Teaching and Learning Programming
University of Notre Dame
2023-2025
Vanderbilt University
2023-2024
University of Michigan
2021-2023
Google (United States)
2023
Recent language models have demonstrated proficiency in summarizing source code. However, as many other domains of machine learning, code lack sufficient explainability --- informally, we a formulaic or intuitive understanding what and how learn from Explainability can be partially provided if, the to produce higher-quality summaries, they also align deeming same parts important those identified by human programmers. In this paper, report negative results our investigation summarization...
Neural code summarization leverages deep learning models to automatically generate brief natural language summaries of snippets. The development Transformer has led extensive use attention during model design. While existing work primarily and almost exclusively focused on static properties source related structural representations like the Abstract Syntax Tree (AST), few studies have considered human attention, that is, where programmers focus while examining comprehending code. In this...
Source code summarization is the task of writing natural language descriptions source code. The primary use these in documentation for programmers. Automatic generation a high value research target due to time cost programmers themselves. In recent years, confluence software engineering and artificial intelligence has made inroads into automatic through applications neural models that However, an Achilles’ heel vast majority approaches they tend rely solely on context provided by being...
Understanding how novices reason about coding at a neurological level has implications for training the next generation of software engineers. In recent years, medical imaging been increasingly employed to investigate patterns neural activity associated with activity. However, such studies have focused on advanced undergraduates and professionals. human study 31 participants, we use functional near-infrared spectroscopy measure introductory programming. controlled, contrast-based experiment,...
Purpose: The fine-tuning of linguistic prosody in later childhood is poorly understood, and its neurological processing even less well studied. In particular, it unknown if grammatical left- or right-lateralized versus adulthood how phonological working memory might modulate such lateralization. Furthermore, virtually develops neurologically among children with cochlear implants (CIs). Method: Normal-hearing (NH) ages 6–12 years NH adults 18–28 completed a functional near-infrared...
Abstract Alzheimer’s disease (AD) is a highly heterogeneous neurodegenerative condition. The current study identified clinically relevant molecular subtypes of the early and late mild cognitive impairment (EMCI LMCI) stages AD using 401 patients’ data from ADNI consortium. We integrated metabolomics with PBMC transcriptomics an unsupervised clustering method called Similarity Network Fusion (SNF), two in MCI patients, respectively. differences between these subtypes’ metabolite...
Medical imaging studies of software engineering have risen in popularity and may reveal the neural underpinnings coding activities. To date, however, all computer science venues treated brain regions independently isolation. Since most complex activity involves coordination among multiple regions, previous analyses overlook behavior.
Code summarization is the task of creating short, natural language descriptions source code. It an important part code comprehension and a powerful method documentation. Previous work has made progress in identifying where programmers focus as they write their own summaries (i.e., Writing). However, there currently gap studying programmers’ attention read with pre-written Reading). As result, it unknown how these two forms compare: Reading Writing. Also, limited understanding programmer...
Abridged: Programmer attention represents the visual focus of programmers on parts source code in pursuit programming tasks. We conducted an in-depth human study with XY Java programmers, where each programmer generated summaries for 40 methods from five large projects over one-hour sessions. used eye-tracking equipment to map while they wrote summaries. also rate quality summary. found eye-gaze patterns and metrics that define common behaviors between during context-aware summarization....
Formal methods are used successfully in high-assurance software, but they require rigorous mathematical and logical training that practitioners often lack. As such, integrating formal into software has been associated with numerous challenges. While educators have placed emphasis on formalisms undergraduate theory courses, such courses struggle poor student outcomes satisfaction. In this paper, we present a controlled eye-tracking human study (n = 34) investigating the problem-solving...
This paper launches a new effort at modeling programmer attention by predicting eye movement scanpaths. Programmer refers to what information people intake when performing programming tasks. Models of refer machine prediction is important people. are because they help researchers build better interfaces, assistive technologies, and more human-like AI. For many years, in SE have built these models based on features such as mouse clicks, key logging, IDE interactions. Yet the holy grail this...
This paper launches a new effort at modeling programmer attention by predicting eye movement scanpaths. Programmer refers to what information people intake when performing programming tasks. Models of refer machine prediction is important people. are because they help researchers build better interfaces, assistive technologies, and more human-like AI. For many years, in SE have built these models based on features such as mouse clicks, key logging, IDE interactions. Yet the holy grail this...
Understanding how novices reason about coding at a neurological level has implications for training the next generation of software engineers. In recent years, medical imaging been increasingly employed to investigate patterns neural activity associated with activity. However, such studies have focused on advanced undergraduates and professionals. human study 31 participants, we use functional near-infrared spectroscopy measure introductory programming. controlled, contrast-based experiment,...