Overview
This app walks you through a complete data analysis pipeline using a synthetic dataset. We simulate a product-based dataset and run in-depth analysis across 20 metrics, before diving into a layer of sentiment analysis per product. Every step is interactive and visualized through clean, informative charts.
Creating the Dataset
We generate a synthetic dataset that includes a variety of fields such as Product Name, Manufacturer, Price, Category, and Stock. This diverse, randomized data simulates real-world conditions and is ideal for analytical practice.
Analyzing the Data
We begin by analyzing the dataset across 20 different metrics. These include checks on data quality, distributions, value ranges, frequency of values, correlations, and more. For each metric, visual insights are provided via bar charts, line plots, and histograms.
Sentiment Analysis
The next step involves sentiment scoring for each product using a simulated sentiment engine. We calculate average sentiment scores and categorize products based on sentiment bands ranging from very negative to very positive.
Visualizing Sentiment Insights
From the sentiment data, we extract deeper insights:
- Top and bottom sentiment-scored products
- Average sentiment per manufacturer
- Sentiment distributions across defined bands
- Best and worst sentiment range per brand
Why Use This App?
This tool demonstrates how to manage and analyze datasets with multiple layers of complexity — from raw numbers to subjective sentiment.
Scalability
While this example uses synthetic data, the same structure can be applied to real-world product datasets — scaled up, connected to APIs, or integrated with actual reviews and ratings.