TikTok Trend Analysis With Python: A Story About CreepyTok
How a Python pipeline turns TikTok search results, metadata, captions and transcripts into a reusable ChatGPT handoff for analysing trends and generating new video concepts.
Tag
research-methods
Research pipeline
A multi-stage constraint system that reconstructs, filters, and stress-tests a search-space to identify which semantic structures are stable enough to act on.
Clustered Instagram Output: Model-Generated Symbolic Compression
This page preserves a model-generated symbolic compression of nine clusters derived from Instagram image-analysis data. It is an interpreted intermediate record, not the raw dataset or a measurement of audience response.
What the Data Actually Looks Like
This post documents the intermediate data produced after scraping and initial structuring. It shows what is actually being analysed before any creative conclusions are formed.
Mining Google Maps Reviews for Unmet Customer Demand
A JupyterLab workflow for extracting Google Maps reviews, filtering for causally useful customer experience signals, and turning local competitor reviews into unmet-demand analysis.
Semantic SEO in Python: From Search Results to Original Content Strategy
How Python can be used to collect search-result data, expose semantic structure, identify unresolved topic clusters, and develop original content strategy from evidence rather than imitation.
Python for NLP and Semantic SEO: A Practical Reference
A plain-language companion to the SEO Python codebase, explaining how the workflow uses search-result collection, NLP text processing, keyphrase extraction, embeddings, network analysis, and association rule mining for semantic SEO research.