📖 Life of Introspection - Personal NLP Research Project 📖

By: Maya Soleimani

Summer I&II - University of Michigan - Dearborn - Professor Maxim

This project parses through 300 diary entries in a 10 year range, capable of calculating different statistics of your choosing of emotional polarity vs. specified independent variables.

Requirements:

--> Written in Python

Libraries used:

--> import datetime

--> from matplotlib import pyplot as plt

--> from nltk.sentiment import SentimentIntensityAnalyzer

Files Needed:

--> loi.py

--> .txt file with your entries where each line follows this format:

“05/17/2014 1853 Today was such a bad day, and I’m just not feeling well overall. For some reason, it’s been so hard to wake up in the morning. 1906”

Extracted: [date, start time, entry, end time]

--> .txt file with manually labeled polarity score with entry to test accuracy. Amount of lines should be a decent chunk of original entries, in this project I used approximately 1/10th of original data. The format of this file follows this format:

"negative Today was such a bad day, and I’m just not feeling well overall. For some reason, it’s been so hard to wake up in the morning."

Instructions:

• In this section, you may type: 'Year range','Winter', 'Summer', 'Spring', 'Fall', 'Words', 'Started', OR 'Duration' without quotations marks.

• Let say you've chosen the year range 2016-2019, you will be then asked for two events A & B.

• Here you will enter input based on what the options say. After that you will receive a bayes probability and a plot of polarity vs. the indepdent variable you selected previously.

• Boom! You have instant insight results on your emotional polarity, based on your personal writing!

Current known errors:

N/A