# HN mention sentiment: praise, critique and reading intent Audit sentiment toward each named book or entity in pasted JSON or CSV. Separate praise, critique, reading intent and neutral mentions before counting recommendations. > Canonical page: https://elysiatools.com/en/tools/hn-mention-sentiment-positive-negative-classifier - **Category:** AI Tools - **Keywords:** HN, mention sentiment, entity sentiment, book recommendations, praise, critique, reading list ## Overview Paste up to 20 rows with entity and context; optional id and commentId preserve traceability. Context must contain the entity literally. Hybrid mode sends only unresolved contexts to OpenRouter using deepseek/deepseek-v4.1-flash-1017 in one request (45 seconds, no retries or model substitution). Rules-only mode makes no model calls. Scores are heuristic or model-reported, not calibrated accuracy; models can err. Mixed, low-confidence and unresolved mentions never vote automatically. Distinct mentions are not recommendations: accepted praise is deduplicated by commentId/entity or exact entity/context. Aliases are not merged. Download JSON, formula-safe CSV and an unconfirmed correction template. Apply corrections only to the same input by preserving id and fingerprint and setting confirmed: true. This tool does not build a persistent annotation database or train models. Platform and model-provider retention policies still apply. ## Inputs - **Mentions (JSON or CSV)** (textarea) - **Input format** (select) - **Classification mode** (select) - **Confirmed corrections (JSON, optional)** (textarea) - **Minimum praise confidence** (number) ## When to use - Compiling curated recommendation lists from forum threads without counting negative reviews as endorsements. - Distinguishing expressions of reading intent or to-read backlog mentions from verified positive praise. - Auditing developer sentiment toward libraries or tools across batch comment excerpts with full traceability. ## How it works - Paste up to 20 rows in JSON or CSV format with required entity and context fields, plus optional id and commentId attributes. - Choose between deterministic Rules-only classification or Hybrid mode, which delegates unresolved contexts to a single external model request. - Evaluate extracted sentiment against a praise confidence threshold to tally deduplicated recommendations and export results as JSON or formula-safe CSV. ## Use cases - Community book list curation where expressions like 'I hated it' must be separated from actual endorsements. - Developer sentiment auditing across technical discussion threads mentioning competing frameworks or tools. - Literature review preparation to isolate acquired reading intent from completed, highly rated titles. ## Frequently asked questions ### Why are total mentions not equal to recommendation counts? A mention may express hatred, balanced criticism, or merely an intention to read, so only deduplicated mentions meeting the praise confidence threshold count as recommendations. ### What is the difference between Rules-only and Hybrid mode? Rules-only mode evaluates patterns locally without external network requests, whereas Hybrid mode submits ambiguous or unresolved contexts to a single fixed model call. ### Why must the context string contain the entity literally? Literal matching ensures the sentiment classifier anchors its evaluation to the exact named subject rather than surrounding unrelated clauses. ### How does the tool prevent duplicate votes? Accepted praise is deduplicated using matching commentId and entity pairs, or identical entity and context pairings when commentId is omitted. ### How do I apply manual corrections to my results? Download the unconfirmed correction template, update the classifications, keep the generated id and fingerprint intact, set confirmed to true, and paste the JSON into the corrections input. ## Related tools - [Structured Log Analyzer](https://elysiatools.com/en/tools/structured-log-analyzer): Detect common log formats, extract core fields, infer field types, and export parsed logs as JSON, CSV, or SQL inserts - [PDF Table Extractor to CSV/JSON](https://elysiatools.com/en/tools/pdf-table-extractor-to-csv-json): Extract tables from PDFs with OpenDataLoader and export them as structured JSON, flat CSV, or HTML tables - [Dataset Imbalance Detector & Resampler](https://elysiatools.com/en/tools/dataset-imbalance-detector-resampler): Detect class imbalance in CSV or JSON datasets, compare resampling strategies, and preview a balanced output dataset - [Time Series Forecast & Seasonality Analyzer](https://elysiatools.com/en/tools/time-series-forecast-seasonality-analyzer): Forecast future periods from CSV or JSON time-series data and inspect trend, seasonal, and residual decomposition in one report - [Train/Test Split with Stratification](https://elysiatools.com/en/tools/train-test-split-with-stratification): Class-stratified train/validation/test split or stratified k-fold for CSV/JSON datasets — seeded, reproducible, with distribution reports, leakage checks, SMOTE preview and CSV export. - [Audio Voice Activity Segmentation](https://elysiatools.com/en/tools/audio-voice-activity-segmentation): Detect likely speech-range activity and export speech/non-speech ranges as JSON, CSV, SRT markers, or an FFmpeg cut list. - [Messy Text to Structured Data Workbench](https://elysiatools.com/en/tools/messy-text-to-structured-data-workbench): Infer tables from pasted text, mixed delimiters, key-value lines, or HTML and export CSV, JSON, and Markdown. - [Mock Data Prefix / Abbreviation Conflict Detector](https://elysiatools.com/en/tools/mock-data-naming-conflict-detector): Detect visually confusing field names and prefix collisions in CSV, JSON, or schema inputs, then suggest clearer renames ## Samples - [CSV Samples](https://elysiatools.com/en/samples/csv-samples): Sample CSV files with various data types, sizes, and complexity levels - [Python Samples](https://elysiatools.com/en/samples/python): Essential Python code examples and Hello World demonstrations - [Distributed Tracing Samples](https://elysiatools.com/en/samples/distributed-tracing-samples): Comprehensive distributed tracing examples using Jaeger, OpenTelemetry, and other modern observability tools for microservices architecture - [AWS EventBridge Samples](https://elysiatools.com/en/samples/eventbridge-samples): AWS EventBridge examples including event buses, rules, targets, schema registry, custom events, and cross-account event routing for serverless event-driven architecture ## Related content - [HN Entity Recommendation Review and Ranking Handoff](https://elysiatools.com/en/hubs/hn-entity-recommendation-review-and-ranking-handoff): Turn an authorized HN comment export into a traceable recommendation table: preserve context, separate praise from critique and reading intent, review disputed votes and deliver reconciled counts.