Member Avatar for tondy

What are good research topics for computer science degree students. Am interested in doing a research in viruses but do not know a good topic to center my research on. Any research topic is welcome even if it is not about viruses.

  • 3 Contributors
  • 4 Replies
  • 187 Views
  • 6 Days Discussion Span
  • Latest Post Latest Post by johnsteve.bravo

Dani AI

Generated

A practical way into a CS research project on viruses is to pick one narrow, measurable question that matches available tools, data and supervisor support. Since asked about virus topics and and pointed toward AI and security angles, projects that combine hands-on analysis with a clear evaluation plan are especially productive. Any experiments with real malware must run in isolated VMs or sandboxes and have supervisor and institutional approval.

  • Binary-as-image classification (CV + malware): map executables to images and train a CNN; compare to a simple static-feature baseline using accuracy, precision/recall and saliency maps for explanations.
  • Runtime-sequence modeling: extract API/system-call traces from a controlled sandbox and model them with sequence learners to detect polymorphic or memory-resident threats; evaluate detection latency and false-alarm rate.
  • Packer/obfuscator detection and unpacking: detect packed samples, measure unpack success, and quantify how unpacking improves downstream analysis.
  • Ransomware early-detection + rollback: detect encryption patterns quickly and test snapshot/rollback recovery; measure time-to-detect and recovery success.
  • Lightweight IoT/firmware detection: design features/models that run on constrained hardware and evaluate cross-architecture generalization.
  • Family attribution and code-similarity: build similarity measures (graph or embedding-based) to cluster/attribute samples and report clustering purity and attribution accuracy.

Project design notes: prefer public, well-labeled corpora where possible; define a simple baseline; use clear metrics (precision, recall, F1, AUC, detection time); run cross-validation or held-out testing; document environment and publish code for reproducibility. Safety and legality are essential: never run samples on production networks, keep samples offline, and follow institutional policies. A good thesis-sized scope is one concrete detection or analysis method with thorough evaluation and clear comparisons to a baseline.

Recommended Answers

All 4 Replies

Member Avatar for mamdouhweb

AI, is a new topic ( comparing it with other topics in CS), so you have a lot of topics like data mining, NLP, Neural Networks, Expert Systems.....
but what the word need in researche the most is how to get the computer to understand a human... so you can follow on, in NLP like isolated words recognition in some language, or more complex like CSR (continues speech recognition). btw i'm working on my final project in CS and i'm having isolated words recognition in android system as its topic

we can work together if you want :D.... Regards..

Member Avatar for tondy

thank you for the ideas, will check with my supervisor

Member Avatar for mamdouhweb

you are welcome, and let me know if i helped you enough

Member Avatar for johnsteve.bravo

You can conduct research regarding security, backup, software development models, quality assurance practices and artificial intelligence.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.