Performance Measurement of Multiple Supervised Learning Algorithms for Gender Identification from Bengali Names

Gender recognition of names is today regarded as one of the most significant problems in data extraction. Machine learning and NLP are used to solve this. Recognition of the gender of a name relies heavily on various kinds of NLP applications such as Question addressing systems, Text summarization etc. Throughout this article, we will investigate the Bengali name entity and attempt to extract and recognize the Person's name entity to determine if the name is male or female. Most people's names have a sexual orientation refinement. However, determining genders from Bengali names with greater accuracy might be difficult. We will demonstrate an artificial intelligence-based characterization technique that can accurately determine sexual orientations from Bengali names. We'll use a variety of algorithms to figure out which computations get the best results. We attained an accuracy of 84 percent by using Naive Bayes.

I. Introduction

For several research questions, demographic information like a person's gender is essential. However, obtaining this type of information is difficult. As a result, in this study, we employed a variety of algorithms to infer genders from Bengali names. The information contained in a person's name is extensive. The process of naming a person is influenced by a person's religion, traditions, beliefs, and geographic region. Some names have a deeper meaning in Bangladesh. A few names are descended from a previous generation. As well as certain names are completely meaningless. There are several names that are used for both males and females. Muslims make up the majority of Bangladesh's population. As a result, Muslim names outnumber those of other religions. The majority of Muslim names are derived from Arabic languages. Other religious groups' names are determined by their beliefs and tradition. Also, certain criteria may be used to determine a person's gender.