Text Data Cleaning and Keyword Extraction
Process text records into usable categories.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Clean messy text records and extract useful keywords or categories. Language and libraries: [LANGUAGE_AND_LIBRARIES]. Sample records (include messy ones): [PASTE_SAMPLE_RECORDS]. Goal: [GOAL] (such as categorizing support requests). Languages in the text: [TEXT_LANGUAGES]. Show the work as BEFORE and AFTER for the sample records. BEFORE: five sample records as they are, with problems marked in square brackets, for example [html tags], [mixed case], [extra whitespace], [emoji], [misspelling], [signature block], [duplicate text], [personal data]. AFTER: the cleaned record produced by your pipeline, and the keywords and category assigned, with the reasoning for each. Then deliver the code: cleaning functions (normalization, removing markup, handling punctuation and numbers, optional spelling normalization, de-identification of emails and phone numbers), tokenization with stop word handling suited to the language, keyword extraction (frequency and TF-IDF), phrase detection, rule-based categories with a keyword dictionary I can edit, and an evaluation: label 50 records by hand and compare. Provide a review file output for uncertain cases. Warn about pitfalls: over-cleaning that removes meaning, language mixing, and bias in categories. Keep personal data out of logs.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.