Synthetic data is generated algorithmically rather than collected from the underlying real-world process. Its cleanliness and uniform structure can speed up prototypes, but synthetic distributions may not represent real markets or preserve useful ground truth. section 4.1
Use synthetic data selectively:
- Choose it for mock data, demonstrations, algorithm education, and testing machine-learning workflows. section 4.1
- Treat it cautiously for exploratory analysis, market studies, demand forecasting, or other tasks requiring real-world distributions. section 4.1
- Check whether generated fields have meaningful semantics, realistic distributions, and ground truth before treating outputs as actionable. section 4.1
Medicine datasets
A synthetic medicine dataset with about 50,000 entries may be cleaner and easier for controlled model demonstrations. A scraped Indian medicine dataset with about 253,973 records is harder to clean but better suited to real pricing, manufacturer, composition, and market analysis. The choice follows the question: prototype the workflow with synthetic data; study the pharmaceutical market with real data. section 4.1
|
Use Case
|
Dataset A (Indian)
|
Dataset B (Synthetic)
|
|
Exploratory Data Analysis (EDA)
|
✅ Strong — real
distributions, price trends
|
⚠️ Limited — synthetic distributions
|
|
Supervised ML for Prediction
|
⚠️ Limited (no target label by
default)
|
⚠️ Same — likely no target labels
|
|
Clustering / Unsupervised learning
|
✅ Yes — cluster by
manufacturer, composition
|
⚠️ Yes but less real-world
meaning
|
|
NLP/Categorization
|
⚠️ Yes — text fields (names/composition)
|
⚠️ Yes (names) but synthetic
semantics
|
|
Recommender systems
|
⚠️ Possible but needs
augmentation
|
❌ Limited
|
|
Benchmarking models / DEMO
|
⚠️ Possible but raw
|
✅ Ideal due to cleaner,
controlled data
|
|
Aspect
|
Indian Medicine Dataset (A)
|
Kaggle Medicine Dataset (B)
|
|
Source
|
Scraped real data
|
Synthetic generated data
|
|
Size
|
~250 K records
|
~50 K records
|
|
Real-world
|
✅ Yes
|
❌ No
|
|
Cleanliness
|
⚠️ No (requires cleaning)
|
✅ Likely
|
|
Usability for Research
|
High
|
Moderate
|
|
ML Power
|
Good for real analysis
|
Good for ML practice
|
Compare synthetic and scraped data across exploratory analysis, prediction, clustering, NLP, recommender systems, benchmarking, size, cleanliness, and research value. section 4.1























