40 computer science research project ideas for high school

Choosing the right research project is the first real decision of your academic career. These 40 computer science research project ideas for high school give you a starting point that is rigorous, original, and publishable.

Computer science is one of the most accessible fields for high school researchers. You do not need a laboratory. You do not need expensive equipment. You need a clear question, a methodical approach, and the discipline to follow through. The ideas below span artificial intelligence, cybersecurity, data science, human-computer interaction, and more. Every one of them is scoped for a high school student working independently or with a faculty mentor.

Why Computer Science Research Matters at the High School Level

Original research in computer science is not reserved for university graduate students. High school students are already building apps, analyzing datasets, and training machine learning models. The gap between a personal project and a published research paper is smaller than most students think. What separates the two is structure: a defined research question, a reproducible methodology, and a clear contribution to existing knowledge.

Publishing your work gives it permanence. A DOI-assigned paper exists in the academic record indefinitely, findable by researchers, admissions officers, and future collaborators. If you want to understand what that process looks like in practice, our guide on how to write a computer science research paper in high school walks through every stage from abstract to references.

40 Computer Science Research Project Ideas for High School

The list below is organized by subfield. Each idea is framed as a researchable question or project direction, not just a topic label. Pick the one that matches your existing skills and genuine curiosity.

Artificial Intelligence and Machine Learning

1. Bias in facial recognition systems: Test a publicly available facial recognition API across demographic groups and document accuracy disparities.

2. Sentiment analysis of student-written text: Train a model on anonymized student essays and evaluate whether sentiment correlates with teacher-assigned grades.

3. Predicting academic performance: Use publicly available educational datasets to build a classification model that identifies at-risk students based on attendance and assignment completion patterns.

4. AI-generated vs. human-written text detection: Build or evaluate a classifier that distinguishes between GPT-generated and student-written paragraphs.

5. Transfer learning for low-resource languages: Fine-tune a pre-trained NLP model on a language with limited training data and measure performance degradation.

6. Reinforcement learning in simple game environments: Implement a Q-learning agent in a grid-based game and analyze how reward structure affects convergence speed.

7. Image classification under adversarial conditions: Introduce pixel-level perturbations to a dataset and measure how much accuracy drops on a convolutional neural network.

Cybersecurity and Privacy

8. Password strength and user behavior: Survey a sample population on password practices and analyze the relationship between security literacy and password complexity.

9. Phishing email detection using NLP: Build a classifier trained on labeled phishing datasets and compare performance across different feature extraction methods.

10. Privacy policy readability analysis: Apply readability scoring algorithms to the privacy policies of major platforms and correlate complexity with user data collection scope.

11. Vulnerability patterns in open-source code: Analyze a public repository of known CVEs to identify which programming languages or code patterns appear most frequently in security disclosures.

12. Two-factor authentication adoption barriers: Conduct a structured survey on why users disable 2FA and model the behavioral factors involved.

Data Science and Statistical Computing

13. Climate data trend analysis: Use NOAA or NASA public datasets to model temperature anomalies at the local or regional level over a 50-year period.

14. Social media language and mental health indicators: Analyze public Twitter or Reddit datasets for linguistic markers associated with self-reported anxiety or depression.

15. Sports performance prediction models: Build a regression model using historical player statistics to predict future performance metrics in a sport of your choice.

16. Economic inequality and internet access: Use World Bank and ITU datasets to model the relationship between broadband penetration and GDP per capita across countries.

17. Algorithmic bias in hiring tools: Analyze publicly available audit studies of resume-screening algorithms and synthesize findings into a structured literature review with original statistical commentary.

18. Traffic pattern prediction using open city data: Use municipal open-data portals to build a time-series model predicting peak congestion windows.

Data analysis is the backbone of most of these projects. If you need a structured approach to handling datasets at the high school level, our resource on how to analyze data in a high school research project covers cleaning, visualization, and statistical interpretation.

Human-Computer Interaction

19. Accessibility of educational software for students with dyslexia: Evaluate three or more widely used learning platforms against WCAG accessibility standards and user testing with affected students.

20. Dark patterns in mobile app design: Catalog and categorize manipulative UI patterns in a sample of top-downloaded apps and assess their frequency by app category.

21. Notification overload and cognitive performance: Design a controlled study measuring how smartphone notification frequency affects task completion time and error rate.

22. Voice assistant accuracy across accents: Test a commercial voice assistant on standardized prompts read by speakers with different regional accents and measure word error rates.

23. Color contrast and reading comprehension: Run a user study comparing reading speed and retention across different foreground-background color combinations in digital text.

Algorithms and Computational Theory

24. Comparative analysis of sorting algorithms on real-world data: Benchmark standard sorting algorithms on datasets with different distributions and document where theoretical complexity diverges from practical performance.

25. Graph algorithms for social network analysis: Apply centrality measures to a public social network dataset and analyze which nodes function as information bottlenecks.

26. Pathfinding algorithms in dynamic environments: Compare A* and Dijkstra performance in grid environments where obstacles change mid-execution.

27. Compression algorithm efficiency on different file types: Benchmark lossless compression algorithms across text, image, and audio file categories and analyze trade-offs between compression ratio and speed.

Computer Science Education and Equity

28. Gender representation in high school CS curricula: Analyze AP Computer Science enrollment data by gender across school districts and correlate with school demographic and funding variables.

29. Effectiveness of block-based vs. text-based programming for beginners: Conduct a study comparing learning outcomes for students introduced to programming through Scratch versus Python.

30. Rural vs. urban access to CS education: Use public school data to map disparities in CS course availability by geography and propose an evidence-based intervention model.

31. Peer instruction in introductory programming courses: Design and evaluate a peer-tutoring protocol for a CS club or class and measure its effect on quiz scores and self-efficacy.

Interdisciplinary and Applied Projects

32. Machine learning for early disease detection: Use a public medical imaging dataset (such as chest X-rays) to train and evaluate a binary classifier for a specific condition.

33. Natural language processing for legal document summarization: Build a summarization pipeline for publicly available court opinions and evaluate output quality against human-written summaries.

34. Computational modeling of misinformation spread: Simulate information diffusion on a network graph and test how different intervention strategies (flagging, removal, counter-messaging) affect spread rate.

35. Algorithmic music generation: Train a recurrent neural network on MIDI data from a specific genre and evaluate output quality using both automated metrics and human listener ratings.

36. Computer vision for plant disease identification: Train an image classifier on a labeled plant disease dataset and assess its viability as a low-cost agricultural diagnostic tool.

37. Predictive modeling for energy consumption: Use smart meter data from public utility datasets to build a model predicting household energy use by time of day and season.

38. Chatbot effectiveness in mental health support: Review and synthesize the existing literature on AI-based mental health chatbots, then design an original evaluation framework for assessing response quality.

39. Code readability and collaboration outcomes: Analyze open-source GitHub repositories to test whether projects with higher code readability scores (measured by automated tools) attract more contributors.

40. Automated fact-checking pipeline design: Build a prototype system that cross-references claims extracted from news articles against structured knowledge bases and evaluate precision and recall.

How to Choose the Right Project from This List

Forty options can feel overwhelming. Narrow the list using three filters. First, ask what data or tools you already have access to (a project requiring proprietary hospital data is not viable without institutional partnership). Second, ask whether the question is specific enough to answer within a defined scope (broad topics become publishable papers only when bounded by a clear hypothesis or research question). Third, ask whether you can articulate why the answer matters (reviewers and readers want to know the contribution, not just the method).

If your project started as a science fair entry, it may already have the empirical foundation for a full paper. Our guide on how to turn a science fair project into a published research paper explains how to make that transition without starting from scratch.

Connecting CS Research to Other Disciplines

Several ideas on this list sit at the intersection of computer science and other fields. That is not a weakness; it is a strength. Interdisciplinary research often produces the most original contributions because it applies computational methods to questions that domain specialists have not yet solved algorithmically. If your interests extend beyond pure CS, explore how computation intersects with behavioral science research in high school or how data analysis applies to environmental science research. The methodological skills transfer directly.

Economics is another productive intersection. Computational economics and algorithmic market analysis are active research areas. If that direction interests you, our collection of high school economics research paper ideas includes several that pair well with data science methods.

Where to Publish Your Computer Science Research

Completing a project is only half the work. Publishing it is what transforms a school assignment into a credential. Not every journal accepts high school authors, and not every journal that does accepts them seriously (no shortcuts, no rubber stamps). You want a venue with genuine peer review, editorial standards that match university-level expectations, and a DOI that makes your work permanently citable.

If you are evaluating your options, our breakdown of the best computer science journals for high school students compares the major venues by review process, scope, and accessibility. Read it before you submit anywhere.

Princeton JPCR publishes original research across all academic disciplines, including computer science and its interdisciplinary applications. Review is double-blind (your school name and country do not influence the outcome). Every accepted paper receives a DOI. The standard is the work itself, nothing else.

Conclusion

These 40 computer science research project ideas for high school are a starting point, not a ceiling. The best project is the one you pursue with genuine rigor and honest curiosity. Pick a question that you cannot stop thinking about, build a methodology you can defend, and produce work that adds something real to the conversation. That is what exceptional student research looks like, and it deserves an exceptional platform. Explore more research guidance across disciplines on our Blogs page, and when your work is ready, submit it to a journal that takes it as seriously as you do.