Student Posters and Data Science ChallengeISEC 2025

Data Science Challenge

In 2025, ISEC will host its Student Data Challenge Competition (SDC). SDC provides a unique platform for undergraduate, graduate (Masters or early PhD), aspiring data scientists, and analysts to showcase their abilities and compete against their peers. The competition is hosted in conjunction with the ISEC conference, offering participants the opportunity to network with industry experts, gain valuable insights, and receive recognition for their achievements. The SDC challenges students to tackle real-world data-driven problems using their knowledge of data analysis, machine learning, and statistical modeling.

The Challenge

The data challenge is based on predicting software bug report priority (P1-P5) using machine learning (ML) or deep learning (DL) aims to develop models that can automatically assess and prioritize bug reports based on their content. Participants are required to build predictive algorithms that analyze various features of bug reports, such as descriptions, error messages, and historical data, to determine the urgency and impact of each bug. The challenge emphasizes the application of advanced ML and DL techniques, including natural language processing (NLP) and neural networks, to enhance the accuracy of prioritization. Successful models should effectively differentiate between high and low-priority issues, enabling more efficient bug management and resolution processes in software development.

In this competition, you’ll gain access to two similar datasets that include bug report information like issue-id, component, title, description, status, etc. One dataset is titled train.csv, and the other is titled test.csv.

Train.csv will contain the details of 52995 bug reports and bug priority (P1-P5), also known as the “ground truth”.

The test.csv dataset contains similar information but does not disclose the “ground truth” for the bug report. It’s your job to predict these priorities.

Using the patterns you find in the train.csv data, predict the bug report priority of 1329 reports.

Submissions

How to Submit Your Prediction

The competition will be hosted on the Kaggle platform. All the details, such as Data and instructions, will be available on the Kaggle platform.

Submission File Format

You should submit a .csv file with exactly 1329 entries plus a header row. Your submission will show an error if you have extra columns or rows.

The file should have exactly 2 columns:

• Issue-ID (sorted in any order)
• Predicted Priority (P1-P5)

Evaluation Criteria

Your job is to predict a priority label out of P1, P2, P3, P4, and P5 for each bug report (Issue-ID). For each in the test set, you must predict a label between P1 and P5.

Metric

Macro F1-score will be used as the evaluation metric.

Poster Session at ISEC 2025

In the poster session, the selected top students (Top 10) will have the opportunity to present their work to the judges, who will select three finalists across all categories (graduate/undergraduate/data scientists) based on (a) Approach and research methods, (b) Results and significance contributions, and (e) Quality of the oral and visual presentation.

Categories and Awards

Three winners will be selected overall. The three winners will receive certificates and prizes. The names of the winners will also be posted on the ISEC 2025 website.

Call for Student Posters

ISEC 2025 will host a Student Posters Session, a forum for students to present their research work to the conference attendees. It is an excellent opportunity for students to get feedback, learn how to clearly communicate their research, and network with the leading experts in the field.

Best poster awards

As a bonus, there will be prizes for the best posters, judged on the basis of significance of the problem and clarity of exposition.

Eligibility

Students who:

• are currently enrolled in an undergraduate or graduate program, OR
• have finished their undergraduate or master's and are currently employed in a temporary research associate or research fellow position

are eligible for participation (in case of uncertainty, please email the chairs).

Submission guidelines

Submission link: https://forms.gle/tMqrS2ZKGSy7AZns9

To participate, submit an extended 2-page abstract. The abstract should clearly specify the problem, why is it important, the proposed solution, and the results, if any. Early-stage ideas are more than welcome; in such cases, an argument for why the solution may work should be provided. The abstracts will be peer-reviewed and selected abstracts will be invited to the poster session. Details on how to prepare the posters will be shared with the acceptance notification.