KenMwaura1/Chat-With-Your-Mpesa

This project encompasses the extraction and cleaning of transactions from Mpesa PDF statements to the development of APIs for integration with the Streamlit Front End. # Chat With Your Mpesa Statements Ever wanted to analyze your Mpesa spending using the power of Large Language Models (LLMs) without sharing your sensitive data? This project encompasses the extraction and cleaning of transactions from Mpesa PDF statements to the development of APIs for integration with the Streamlit Front End. The project is structured around three main components: - Data Extraction & Cleaning: This phase involves the extraction of transactions from Mpesa PDF statements and the subsequent cleaning and wrangling of the data to prepare it for utilization by the Large Language Model (LLM) application. - Python Code Generation with LLMs: As sensitive data is not shared with the LLMs, the development of the API is crucial. This component focuses on constructing an API capable of accepting user queries and converting them into syntactically correct Python statements. - Streamlit UI: This front end will seamlessly interact with the LLM application APIs, providing users with an engaging platform to analyze their Mpesa transactions. ## Data Extraction & Cleaning The data extraction and cleaning process is crucial for preparing the Mpesa transactions for analysis by the Large Language Models (LLMs). The following steps are involved in this process: - Extraction of transactions from Mpesa PDF statements - Cleaning and wrangling of the data - Preparation of the data for utilization by the LLM application - The data extraction and cleaning process is essential for ensuring that the Mpesa transactions are accurately represented and can be effectively analyzed by the LLM application. - Python Code Generation with LLMs - The Python code generation component focuses on constructing an API capable of accepting user queries and converting them into syntactically correct Python statements. This process involves the following steps: - Development of the API for interacting with the LLMs - Conversion of user queries into Python code - Execution of the Pytho …