This repository contains code and data for Sentiment Analysis in Python Financial News Headlines using a Kaggle dataset. The code is organized into a set of tasks that can be run using the pytask task runner.
Author: Ozodbek Ozodov (o.ozodov@outlook.com)
To get started, create the environment with
$ conda env create -f environment.yml
Then activate the environment with
$ conda activate sentiment_analysis_epp
To build the project, type
$ pytask
To run the code in this repository, you will need to have the following software installed on your computer:
- Python 3.10
- pytask
- scikit-learn
- matplotlib
- seaborn
- numpy
- pandas
- nltk
Running the Code After activating the environment, you can run the sentiment analysis tasks using the pytask command in the src directory. There are three task files with .py extensions that perform various steps in the sentiment analysis process.
Once all tasks have been done, a PDF file named paper_py.pdf will be generated in the src directory. If you would like to see the paper without running the code, a copy of the PDF is included in the root directory named paper.pdf.
This project was created with cookiecutter and the cookiecutter-pytask-project template under MIT License.
Special thanks to EPP lecturers Janos Gabler and Hans-Martin von Gaudecker, as well as my friends and colleagues Haris, Luke, Sugarku, and Sharof for sharing ideas and programming experiences.