Bibliographic record
Abstract
With the rapid growth of product review forums, discussion groups, and Blogs, it is almost impossible for a customer to make an informed purchase decision. Different and possibly contradictory opinions written by different reviewers can even make customers more confused. In the last few years, mining customer reviews (opinion mining) has emerged as an interesting new research direction to address this need. One of the interesting problem in opinion mining is Opinion Question Answering (Opinion QA). While traditional QA can only answer factual questions, opinion QA aims to find the authors' sentimental opinions on a specific target. Current opinion QA systems suffers from several weaknesses. The main cause of these weaknesses is that these methods can only answer a question if they find a content similar to the given question in the given documents. As a result, they cannot answer majority questions like "What is the best digital camera?" nor comparative questions, e.g. "Does SamsungY work better than CanonX?". In this paper we address the problem of opinion question answering to answer opinion questions about products by using reviewers' opinions. Our proposed method, called Aspect-based Opinion Question Answering (AQA), support answering of opinion-based questions while improving the weaknesses of current techniques. AQA contains five phases: question analysis, question expansion, high quality review retrieval, subjective sentence extraction, and answer grouping. AQA adopts an opinion mining technique in the preprocessing phase to identify target aspects and estimate their quality. Target aspects are attributes or components of the target product that have been commented on in the review, e.g. 'zoom' and 'battery life' for a digital camera. We conduct experiments on a real life dataset, Epinions.com, demonstrating the improved effectiveness of the AQA in terms of the accuracy of the retrieved answers.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.018 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".