Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2026
|
| Online Access: | https://doi.org/10.5281/zenodo.19541935 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <h3>Abstract</h3> <div>Effective data analytics and machine learning depend heavily on data quality, but real-world datasets frequently have missing values, outliers, duplicate records, and structural irregularities that lower analytical reliability. An AI-based interactive data cleaning and analysis platform that automates the data preparation lifecycle and facilitates effective data exploration is presented in this work. Python and the Streamlit framework are used in the development of the suggested system, which incorporates automated data cleaning methods such as KNN-based imputation for managing missing values and Isolation Forest for anomaly detection. The platform offers an interactive Exploratory Data Analysis (EDA) module that facilitates statistical summaries and visual data exploration in addition to preprocessing. The system includes an analytics interface based on the Large Language Model (LLM) to improve usability. The integration of this method bridges the gap between automated data cleaning and intuitive data interpretation. The experimental use of the system demonstrates that this proposed approach reduces manual preprocessing effort and also improves dataset readiness for downstream analytics and various machine learning tasks.</div> <h3><span class="ez-toc-section"></span>Keywords</h3> <p>Data Cleaning, Artificial Intelligence, Data Preprocessing, Outlier Detection, Missing Value Imputation, Exploratory Data Analysis, Interactive Data Analytics.</p>