Prediction for rating shorts stories of Sherlock Holmes

Regression
Sentiment Analysis
Author

Krishnakanta Maity

Published

June 11, 2023

1 About

Sherlock Holmes is a beloved literary character known for his exceptional detective skills and logical reasoning. The stories featuring Sherlock Holmes have captivated audiences for over a century, making him one of the most popular fictional characters in literature. In this project, we will be conducting a statistical analysis of the Sherlock Holmes short stories to gain a deeper understanding of the sentiments, frequent words, and facts present in the stories.

Our primary objective is to analyze the flow of sentiments throughout the stories. We will be examining the use of language and the emotions conveyed through the words used in the stories. Additionally, we will be identifying the most frequent words and uncovering the sentiment behind them. By gaining a better understanding of the language and emotions present in the stories, we aim to gain a deeper appreciation of the writing style and storytelling techniques used by Sir Arthur Conan Doyle.

Another objective of this project is to uncover the facts present in the stories. We will be analyzing the data to identify the key facts that drive the plot and shape the characters in the stories. This will provide insight into the structure and themes of the stories, as well as the historical and cultural context in which they were written.

Finally, we will be building a model for predicting the rating of the stories. This will provide a deeper understanding of the stories as a whole and how they relate to one another. Overall, this project aims to provide a comprehensive analysis of the Sherlock Holmes short stories and enhance our understanding of the beloved literary character and his stories.

The project aimed to predict ratings for short stories of Sherlock Holmes. The methodology involved using the UDPipe-14 model for feature extraction. The dataset used consisted of 56 short stories written by Arthur Conan Doyle, which were obtained from Kaggle. The findings of the project indicated that the ratings of the stories were primarily influenced by their age and the presence of enthusiastic detective-related words.

2 Source

       

Here, I provide an overview of the project. To delve into the methodology and explore the critical findings, I encourage you to review the accompanying slides and detailed report (above links).

3 Acknowledgment

It is great pleasure for me to undertake this project. I am grateful to my project guide Prof. Tirthankar Ghosh, Department of Statistics, Visva Bharati University, Bolpur, Shantiniketan.