baidu/Senta

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

2015 stars · 363 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Senta is Baidu's open-source sentiment analysis system built on the SKEP (Sentiment Knowledge Enhanced Pre-training) model. It provides pretrained Chinese and English models and one-click tools for sentence-level sentiment classification, aspect-level sentiment classification, and opinion extraction.

Use cases

• classify sentiment of product reviews
• run aspect-based sentiment analysis on Chinese text
• extract opinion targets from reviews
• fine-tune a sentiment pretrained model with SKEP
• deploy a one-click sentiment prediction pipeline
• reproduce ACL 2020 SKEP benchmark results

When to choose

• you need state-of-the-art sentiment analysis for Chinese or English text
• you want pretrained sentiment models with minimal code
• you need aspect-level sentiment classification or opinion extraction
• your stack already uses PaddlePaddle

When to avoid

• you need a PyTorch or TensorFlow-native solution
• you need general-purpose NLP beyond sentiment tasks
• you require actively evolving features, as development has slowed to maintenance

Facets

library · maturity maintenance

nlp machine-learning deep-learning machine-learning artificial-intelligence python cross-platform sentiment-analysis sentiment-classification skep paddlepaddle aspect-level-sentiment opinion-target-extraction pretrained-models chinese-nlp natural-language-processing linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="baidu/Senta")