Can you suggest me some project for computer science?

How do I begin research in theoretical computer science?

• I want to know how to link natural language processing and theoretical computer science.Can you suggest a small project to work on?

• Answer:

  Short Answers : Read. Papers, Journals, Magazines, Thesis, Books, etc. Watch conference videos. Long Answer : I have no idea about your background in NLP or Theoretical CS, but the common start point for any kind of research is reading. Most people generally have misconception that papers are too technical for them, but in reality they are not. Books, especially the curriculum books contain information that was about a decade old. Some traditional books are like 3 - 4 decades old. For getting more information about the current development in a particular field, one has to refer to papers. Initially reading a paper might seem boring and tedious. You might now even understand half the thing. Half the people give up while reading their first paper itself. It took me a week to read my first paper (:P). But with time your pace will increase. Just keep reading and understanding. You can Google technical terms used in the paper or refer to wikipedia page, if any. Make sure you always refer to some authentic source for information. If you have any specific area in mind like NLP, you can directly search for popular journals in that field and start reading at once. If you do not have anything specific in mind, try to explore and find one. I believe the best research can be carried out if you are inspired by a real-life problem. If you are completely blank try reading papers of different field and find something interesting. Force yourself to choose at-least some broad area. Eventually, you will find find your way. For getting started you can browse through the following widely used Digital Libraries : 1. http://dl.acm.org/ 2. http://ieeexplore.ieee.org/Xplore/home.jsp 3. http://link.springer.com/ 4. http://www.sciencedirect.com/ 5. http://www.researchgate.net/ (Networking Website) You can also use Google Scholar for searching academic articles. You also need access rights to some of these libraries. Many Govt Colleges (at least  IITs and NITs) provide free access to popular libraries via LAN/WiFi. Once you find your first interesting paper, the references will take you to related papers and so on. Their will be a kind of chain reaction and you will slowly understand the ideas proposed by various researchers from across the globe. You can even try and email the authors in case you have any question regarding their work. If the question is legitimate, they will reply most of the time. You can take help from your college professors too. And Google is always there for help. After papers, you can move on to Journals and Magazines related to the specific field. Magazines also give you an idea about the current industrial demand, new trends and emerging techniques. This will help you classify work according to their need in the current scenario. After all, necessity is the mother of all inventions :). Thesis are generally very specific and contains an in-depth analysis of a particular topic. After getting a broad idea about the current on-going research in an area, you will start having your own arguments and explanations. One good thing about Theoretical CS research is that there is nothing wrong or right. It's all about arguments, testing and finally determining the "better" way. You can try and implement the idea proposed in a paper and evaluate the results obtained. You can also hybridize two or more independent work to see if you can improve the existing method. You may even propose a completely new idea that you think will be relevant to the context. Trust me, once you start reading a lot about a field, you start getting new ideas automatically. Sometimes to immediately relate o work done in different context to the current context. Slowly things will build up and you will have enough material to write your own paper and that will just be the beginning. Test you proposal, analyze the result and then show off your work to the world. I also think that group work is quite helpful in research because as I said nothing is wrong or right, so multiple opinions about a topic is of utmost importance. Also, Documentation is very very important. I have learned a lot just by roughly documenting whatever I read or find out. Keep a small diary or use sticky notes or use gedit. About the Projects : I think starting a project straight-away is not at all a good idea if you want to get into core research. For taking up good research project you must have some knowledge about the current work. Even if you are doing a project, make sure it is unique and original. You can try and implement various algorithms/mechanisms proposed by other people. Make sure your project contributes significantly to the field. Don't just code something that will never be seen again by anyone else after you stop working. If you badly want to work on a project, I suggest you should try working on a bigger open-source project related to the field - you will learn much more.