Fine-Grained Opinion Mining Using Conditional Random Fields
Bibliographic record
Abstract
User generated data and specially product reviews are major sources of information for costumers to make informed purchase decisions and for producers to keep track of consumers opinions. As e-commerce is becoming more and more popular, the number of customer reviews that each product receives grows rapidly. As a result, the problem of automatically mining reviews to extract useful information has recently attracted many researchers. In the last decade, several works have been presented for identifying product aspects from reviews. Product aspects are components or attributes of the product that have been commented on in the review, e.g. 'zoom' and 'battery life' for a digital camera. In this paper, we propose a novel method for mining user opinions, which aims at extracting not only the opinions of users on product aspects, but also a finer level of information indicating the usage type of the aspect. In other words, we try to find out how the reviewer used the aspect. In this work, we focus on the task of identifying product aspects, corresponding opinions, and related usages as a sequence tagging problem. We employ Conditional Random Fields (CRF) to solve the stated problem and propose techniques for defining and filtering features to enhance the accuracy. The accuracy of the proposed method is evaluated using a real life data set from Epinions.com. Experimental evaluation confirms the improved accuracy of our method in identifying aspects, aspect usages, and related opinions. We also evaluate the effectiveness of the optimization techniques through multiple experiments.
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 | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".