Since the mid-20th century, when artificial intelligence began to emerge as an academic field, a variety of machine learning techniques have been developed. While researchers developed the field, they had difficulties putting these machine learning concepts into practice. Now that the capabilities of computers in our generation have become much more powerful, what had been researched upon has started to be applied to our everyday lives in order to improve our living standards. Even though we are still at the beginning of the age of artificial intelligence and its applications are still far from perfection, they have already reached the point where they are useful. The paper seeks to prove the utility of artificial intelligence in the form of a chatbot, an automated text generator, using a machine learning technique called Long Short-Term Memory, while also analyzing behavioral differences of chatbots learning from datasets consisting of different levels of vocabulary. Conclusively, the results come out to prove the utility of chatbots and behavioral differences to a certain extent.


Sommer, Nathan


Computer Science


Machine Learning, Artificial Intelligence, Chatbot, LSTM, Sequence to Sequence

Publication Date


Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis



© Copyright 2019 Sota Shimizu