Optimizing Artificial Intelligence Algorithms on Data Classification and Generation

Inhalt:
Artificial intelligence has become a hot topic over the past few decades, and scholars tried to apply it to different fields to increase the productivity and precision of experiments or studies. The idea of Algorithm classification and composition has a history of over 60 years, and many commercial software and scholars today adopt artificial intelligence to data classification and generation. In this article, the author, through a method of analyzing the source code and comparing different classification methods like Multilayer Perceptron, Logistic Regression, and Naïve Bayes classifier, will discuss different approaches to digitalize and train the model for composition, as well as analyze the possible flaws of current approaches. Besides, it will propose possible room for improvements for current algorithms to improve their performance. With the analysis of algorithms, the author concludes that the current algorithm still has some drawbacks regarding the lack of ability to recognize different attributes. Algorithm composing still faces some bottlenecks and more in-depth analysis of different labels is needed despite their impressive performance already.