You will get A fully cleaned, dataset with no Inconsistency and zero missing values.
Project details
This project showcases my approach to real-world data cleaning on a
café sales dataset with 10,000 transactions containing dirty placeholder
strings, wrong data types, and up to 40% missing values.
Instead of blindly applying mean/median imputation, I analysed WHY
each column had missing values and wrong data types and chose the correct strategy:
→ Probabilistic sampling for categorical columns (Location, Payment)
→ Business logic arithmetic for numerical columns (Qty × Price = Total)
→ Feature engineering for Transaction Date (Month, DayOfWeek, Quarter)
Result: 100% clean, ML-ready dataset with zero missing values.
Full Jupyter Notebook + Professional PDF report with 12 visualisations
documenting every decision — exactly what I deliver to every client.
café sales dataset with 10,000 transactions containing dirty placeholder
strings, wrong data types, and up to 40% missing values.
Instead of blindly applying mean/median imputation, I analysed WHY
each column had missing values and wrong data types and chose the correct strategy:
→ Probabilistic sampling for categorical columns (Location, Payment)
→ Business logic arithmetic for numerical columns (Qty × Price = Total)
→ Feature engineering for Transaction Date (Month, DayOfWeek, Quarter)
Result: 100% clean, ML-ready dataset with zero missing values.
Full Jupyter Notebook + Professional PDF report with 12 visualisations
documenting every decision — exactly what I deliver to every client.
Data Tool
PythonWhat's included $12
These options are included with the project scope.
$12
- Delivery Time 1 day
- Number of Revisions 3
Frequently asked questions
About Shivam
Data Cleaning, ML Dataset Prep Expert | Python | Pandas | Scikit-learn
Noida, India - 9:53 pm local time
With a background spanning data analytics and applied machine learning — including deep learning-based defect detection systems (PyTorch, CNNs) for manufacturing clients, and campaign performance analytics with dashboards in Tableau and Power BI — I bring practical, real-world experience to every data project, not just theoretical knowledge.
I can help with:
Data cleaning and formatting
Missing value handling and outlier detection
Data preprocessing for machine learning
Exploratory Data Analysis (EDA)
CSV/Excel dataset preparation
Pandas, NumPy, and Scikit-learn-based workflows
I focus on delivering clean, well-documented datasets that are ready for analysis or modeling — with clear communication and on-time delivery every time.
Tools: Python, Pandas, NumPy, Scikit-learn, Jupyter Notebook, Matplotlib, Tableau, Power BI
Currently completing my MSc in Data Science & AI from BITS Pilani, adding academic rigor to hands-on project experience.
Steps for completing your project
After purchasing the project, send requirements so Shivam can start the project.
Delivery time starts when Shivam receives requirements from you.
Shivam works on your project following the steps below.
Revisions may occur after the delivery date.
Data Audit & Analysis
Receive dataset and audit all dirty values, wrong types, and missing patterns
Cleaning & Imputation
Apply column-specific cleaning: replace placeholders, convert types, impute missing values using smart strategies