What it does

This repository is a manually curated index of academic papers, open-source implementations, and datasets focused on sentiment analysis subproblems: emotion cause extraction (figuring out why someone feels a certain way), and opinion target/opinion term extraction (identifying what is being praised or criticized). Most entries link to PDFs; some include GitHub repos or dataset downloads.

The interesting bit

The maintainer has done the tedious work of gathering Chinese-language research from journals like 《中文信息学报》 alongside mainstream ACL/AAAI papers, making this a rare bridge between Chinese NLP scholarship and the usual English-dominated reading lists. The emotion-cause-pair extraction section even links to a live GitHub repo with data.

Key highlights

• ~40 papers across three subfields: sentiment analysis, emotion cause detection, and opinion target extraction
• Mix of English and Chinese sources, including hard-to-find CIPSC journal articles
• Links to practical datasets: SemEval-2014, Baidu/Dianping/Mafengwo reviews, and the ECPE emotion-cause pair corpus
• Sparse but present code links (ECPE, E2E-TBSA, chinese-opinion-target-extraction)
• Last updated roughly 2018–2019 based on paper dates; no explicit maintenance policy stated

Caveats

• No search, tagging, or categorization beyond the three broad sections—finding a specific method means scrolling
• Most entries are paper links only; code and dataset coverage is spotty
• README is entirely in Chinese; English-only readers can navigate paper titles but will miss context

Verdict

Worth bookmarking if you’re doing literature review in emotion cause extraction or Chinese opinion mining, especially for finding non-English sources. Skip it if you need runnable baselines or a systematically maintained benchmark hub.