Hey guys, I've written a study plan for kgsp and I would love your input. I'm planning to apply via uni track in KAIST for a master's degree in Computer Science.
In my bachelors, I was introduced to Artificial Intelligence during a course and I was instantly hooked. It was such a fascinating field for me that it settled the theme of my upcoming research project for me. To pursue my passion of Artificial Intelligence further, I took up the course Natural Language Processing (NLP) and chose to make an NLP based Chatbot for my final year project. While conducting research for that project, I and my group discovered a huge room of improvement in available Chatbots. Most of the Chatbots developed today generate responses based on keyword searching, which after a while proves to be repetitive and unrealistic. That's where NLP came to play, it is a branch of Artificial Intelligence that is concerned with the interactions between computer and human. Intelligent Chatbots are a hot research topic and companies like Google, Microsoft and Facebook are all investing in them.
The main tasks of NLP are syntactic and semantic analysis of the text. During our project's research, we completed the syntactic analysis module but did not get the chance to do the semantic analysis due to time and resources limitations. Semantic Analysis is branch of Natural Language Processing where we deal with the word senses and what that word can mean in a given statement. A word can have one meaning in one context and a totally different one in another. There are many algorithms available for achieving this task, the most popular being the Lesk and Walker's Algorithms and Bayes Theorem, but all of these have some limitations. The success rate of these algorithms is normally 60-65%, which can decrease if we provide large corpus (set of texts). A deep study of these algorithms and to find ways how to improve them is going to be my main focus during my future research.
I also plan to take courses on Virtual and Augmented Reality. These two and my research complement each other. We can create an environment where a person can interact with a virtual being. The inspiration for this idea comes from PokémonGo. In the game we can see the Pokémon characters from the Anime Series, as virtual beings, but how awesome would it be to make them socially interactive. It can be a new step in the world of Chatbots, where we usually converse with them by typing into a text box.
When I'm awarded with this scholarship, I envision to fulfill my aspirations for contributing to, what I believe will be, the next major milestone in the computer field.
Study Plan for graduate school
In my bachelors, I was introduced to Artificial Intelligence during a course and I was instantly hooked. It was such a fascinating field for me that it settled the theme of my upcoming research project for me. To pursue my passion of Artificial Intelligence further, I took up the course Natural Language Processing (NLP) and chose to make an NLP based Chatbot for my final year project. While conducting research for that project, I and my group discovered a huge room of improvement in available Chatbots. Most of the Chatbots developed today generate responses based on keyword searching, which after a while proves to be repetitive and unrealistic. That's where NLP came to play, it is a branch of Artificial Intelligence that is concerned with the interactions between computer and human. Intelligent Chatbots are a hot research topic and companies like Google, Microsoft and Facebook are all investing in them.
The main tasks of NLP are syntactic and semantic analysis of the text. During our project's research, we completed the syntactic analysis module but did not get the chance to do the semantic analysis due to time and resources limitations. Semantic Analysis is branch of Natural Language Processing where we deal with the word senses and what that word can mean in a given statement. A word can have one meaning in one context and a totally different one in another. There are many algorithms available for achieving this task, the most popular being the Lesk and Walker's Algorithms and Bayes Theorem, but all of these have some limitations. The success rate of these algorithms is normally 60-65%, which can decrease if we provide large corpus (set of texts). A deep study of these algorithms and to find ways how to improve them is going to be my main focus during my future research.
I also plan to take courses on Virtual and Augmented Reality. These two and my research complement each other. We can create an environment where a person can interact with a virtual being. The inspiration for this idea comes from PokémonGo. In the game we can see the Pokémon characters from the Anime Series, as virtual beings, but how awesome would it be to make them socially interactive. It can be a new step in the world of Chatbots, where we usually converse with them by typing into a text box.
When I'm awarded with this scholarship, I envision to fulfill my aspirations for contributing to, what I believe will be, the next major milestone in the computer field.
