I’m providing a corpus of research-article text that needs a careful review followed by automated pattern identification. The job is entirely text-based—no numbers or images—so experience with NLP techniques in Python, R, or any mainstream language is important. Scope of work • Parse and clean the article transcripts so they are ready for analysis. • Identify three specific patterns: – Trends over time (how topic prominence changes by publication year) – Keyword frequency (ranking and comparative insights) – Sentiment analysis (positive, negative, neutral tone across the corpus) • Present findings clearly, using concise narrative plus supporting visuals such as time-series plots, frequency charts, and sentiment distribution graphs. Deliverables 1. A cleaned, well-annotated version of the dataset. 2. Script(s) or notebook(s) with reproducible code. 3. An executive summary (PDF or DOCX) that explains methodology, key results, and actionable insights, accompanied by exportable images of the charts. Acceptance criteria • All three pattern categories are covered comprehensively. • Code executes end-to-end on my machine with a single command and includes comments for clarity. • Summary highlights at least five meaningful trends and interprets sentiment shifts in plain language. If this sounds straightforward to you and you have demonstrable NLP or text-mining experience, I’m ready to share a sample of the data so you can confirm approach and timeline.