Picking Strong Computer Science Term Paper Ideas
As computer science and its newer cutting-edge topics such as artificial intelligence continue to advance what it is possible to achieve with the “thinking machines”, there are few other disciplines that are having as profound an effect on all of human life in the modern world. So, there’s no reason to limit yourself when picking ideas for your next term paper… The subject is wide open, with uncountable unanswered questions, and still deeply relevant and exciting. So, sort through the array of lesser topics to find thrilling ideas like these below.
- The NoSQL databases
- Everyone is a computer scientist
- Towards scientific data mining
- Exploring functional languages
- Software: Time to catch up
- Parallel processing
- Cryptography of the future
- Micro-service architectures
- Multi-dimensional clustering at speed
The standard model for databases for decades has been the relational model. Larger datasets and greater demands for concurrent processing have encouraged a slew of new approaches in recent years. Compare and contrast these, and forecast which will stand the test of time.
Computers are becoming as ubiquitous as our most basic tools, such as pen and paper. They’ve been a huge boon to the productivity of almost everyone. Does this mean everyone benefits by becoming computer scientists in their own right?
Harvesting big data for potential insights violates tenets of science to formulate testable hypotheses, and risks simply finding spurious patterns in the noise. How can this weakness be fixed?
The move away from imperative programming promises greater testability and robustness of systems, but the languages themselves are notoriously hard to learn. Can they be made simpler? Or can we teach the more effectively?
Advances in computer hardware have far outstripped advances in our ability to build complex software systems that efficiently use the new hardware. What new ideas are being explored to take the next leap?
As the limits of processors are being reached, parallel computation offers a solution to performance demands. What is the state of the art in terms of building parallelised software?
If an efficient mechanism to factor prime numbers is discovered, modern cryptography will be rendered entirely ineffective. What will replace it?
Small application interfaces can be made available over the Internet. Under what circumstances should they be considered as components of application architectures?
As an unsupervised machine learning technique, clustering is of paramount importance. How can current algorithms achieve orders of magnitude speedups to be useful to new problems?