Motoevsentiment

A package for extracting and analyzing sentiment from EV motorcycle feedback provided by customers and employees.

Source Code

The full source code for this project is available on GitHub.

wsentiment subpackage

This subpackage analyses WhatsApp related messages for sentiment.

Using wsentiment

import by using

from evmotosentiment.wsentiment import wsentiment_score

wsentiment_score takes only string inputs.

• example

wsentiment_score('kofi has lights out')
>> -2

wdataframe

This subpackage changes WhatsApp text export file (.txt) into a dataframe

Using wdataframe

import using

from evmotosentiment.wsentiment import wdataframe

• example

# Suppose you have a text file with data
test_file = "sample.txt"

# Read the text file into a Pandas DataFrame
df = wdataframe(test_file)

# Display the DataFrame
print(df)

              datetime       sender     message
0  2023-09-29 07:49:00    John Doe          ok
1  2023-09-29 08:15:00    Jane Doe  swapped battery

preprocess_text

This subpackage is used to clean text. The cleaning removes words that do not add to the sentiment analysis such as "the,me, etc"

Using preprocess_text

import using

from evmotosentiment.clean_message import preprocess_text

• example

preprocess_text('Turning swap station off at Haatso')
>> 'turning swap station off haatso'

Running using pytest

To run tests in your current Python environment:

pytest

Running Tests with Tox

This project uses Tox to automate testing across multiple Python versions.

Test Environments

The tox.ini is configured to run tests with:

• Python 3.9
• Python 3.10
• Python 3.11

Each environment installs:

• pytest (for running tests)
• pandas (project dependency)

Prerequisites

• Python 3.9, 3.10, and 3.11 installed on your system
• uv for faster installs (optional but recommended)
• Install tox with uv:

uv pip install tox tox-uv