Summit Haque

Summit Haque

Auto-generated profile, based on publicly available information

Summit Haque

Oregon State University, USA

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

17 April 2023 in IEEE Access
Verified
Journal Article
Open access
ZHZahidul Haque
Zahidul Haque
SZSakib Zaman
Sakib Zaman
JSJillur Rahman Saurav
Jillur Rahman Saurav
SHSummit Haque
Summit Haque
SISaiful Islam
Saiful Islam
MAMohammad Ruhul Amin
Mohammad Ruhul Amin
Within the Natural Language Processing (NLP) framework, Named Entity Recognition (NER) is regarded as the basis for extracting key information to understand texts in any language. As Bangla is a highly inflectional, morphologically rich, and resource-scarce language, building a balanced NER corpus with large and diverse entities is a demanding task. However, previously developed Bangla NER systems are limited to recognizing only three familiar entities: person, location, and organization. To address this significant limitation, we introduce a novel Bangla NER dataset B-NER, which was created using 22,144 manually annotated Bangla sentences collected from Bangla newspapers and Bangla Wikipedia. This dataset includes a total of 9,895 unique words which were manually categorized into eight different entity types, such as a person, organization, event, artifact, time indicator, natural phenomenon, geopolitical entity, and geographical location. Inter-annotator agreement experiments were conducted to validate the quality of annotations performed by three annotators, resulting in a Kappa score of 0.82. In this paper, we provide an outline of the annotation guideline illustrated with examples, discuss the B-NER dataset properties, and present benchmark evaluations of the dataset. To establish that B-NER is more comprehensive and balanced in comparison to other publicly accessible datasets, we conducted cross-dataset modeling and validation, i.e. trained NER model on one dataset while tested on another, and found that the model trained on B-NER performed the best in that settings. Furthermore, we performed exhaustive benchmark evaluations based on Bidirectional LSTM with fastText embeddings and sentence transformer models. Among these models, fine-tuned IndicBERT achieved noticeable results with a Macro-F1 of 86%. This dataset and baseline results will be publicly available under a CC-BY 4.0 license in the CoNLL-2002 format to facilitate further research on Bangla NER.


12 October 2026 in ACM Transactions on Computing Education
Verified
Journal Article
Open access
ATAhsun Tariq
Ahsun Tariq
PCPhillip Conrad
Phillip Conrad
CHChristopher Hundhausen
Christopher Hundhausen
SCSanjay Chandrasekaran
Sanjay Chandrasekaran
YQYunhan Qiao
Yunhan Qiao
SHSummit Haque
Summit Haque
Background and Context. Many software development courses involve agile team projects where students are asked to work through a series of sprints toward a final team deliverable. Such team projects often require adding features to or fixing bugs in legacy code bases. A key challenge for students is to distribute the workload so that all students contribute equally to the team project. Previous studies have reported significant variability in individual contributions in team projects, often where one member contributes more than twice the expected share of work. One common approach to assessing the quality of team processes is reflection —predominantly performed during retrospectives after the sprint. Previous work has found that students’ reflections tend to be at surface level, lacking the depth required to make meaningful changes to team processes. This study explores how reflection prompts embedded into team communication channels can help improve the equality of contributions. Objectives. This study investigates whether weekly reflection prompts, delivered through a Slack bot intervention, can improve the equality of contributions in team projects, compared to a Control Group with no reflection prompts. We examine: (1) the impact of reflection prompts on the equality of contributions between the two groups, (2) differences in team retrospectives, including team practices and the depth of reflection, and (3) whether there is a shift in students’ perceptions of their peers in terms of their contributions and interactions. Method. We conducted a between-subjects study of 19 undergraduate software development teams (114 students) enrolled in two offerings of the same course in different academic quarters within a one-year span. In the first quarter (Control Group), automated updates on the current state of individual project contributions were posted daily in the team communication channel. In the second quarter (Experimental Group), additional weekly prompts were added to encourage participation and reflection on the equality of those contributions. We used a mixed-methods approach, combining quantitative analysis of project contributions, qualitative analysis of team retrospectives, and peer evaluation ratings. Findings. The Experimental Group demonstrated significantly less inequality in contributions, with a large effect size for code review inequality ( \(\beta=-0.16\) , \(p<0.01\) , \(f^{2}=0.95\) ) and story point inequality ( \(\beta=-0.07\) , \(p<0.05\) , \(f^{2}=0.38\) ), and medium-large effect size for inequality in the number of issues assigned to each team member, inequality in points awarded for merged pull requests, and inequality in pull requests authored (all \(\beta=-0.10\) , \(p<0.05\) , \(f^{2}=0.34\) ). In retrospectives, a significant difference in reflection depth was observed ( \(\chi^{2}=27.57\) , \(p<0.001\) , df = 3, Cramér's V = 0.15), with the Control Group exhibiting deeper reflection. The groups also differed in focus areas ( \(\chi^{2}=79.35\) , \(p<0.001\) , df = 13, Cramér's V = 0.25), with the Experimental Group focusing more on pull request discussions ( \(p<0.001\) ). Finally, peer evaluations showed higher interaction ratings in the Experimental Group ( \(U=1237.0\) , \(p<0.05\) , Cliff's Delta = \(0.22\) ). Implications. The findings suggest that automated reflection prompts in team communication channels can enhance contribution equality and improve peer interactions, without the need for extensive instructor involvement. These results highlight the potential for lightweight, embedded interventions to foster more equitable collaboration in undergraduate team projects. However, the greater depth of students’ reflections in the Control Group suggests that, in the absence of such an intervention, retrospectives serve as a key venue for team discussions, and that sprint assessment and evaluation may benefit from their structure or from the design of tools that support them. Further replication studies are needed to assess the generalizability of these findings and explore long-term factors that may impact contribution patterns in undergraduate team projects.