Abstract
Evaluation approaches for unsupervised rank-based keyword assignment are nearly as numerous as are the existing systems. The prolific production of each newly used metric (or metric twist) seems to stem from general dissatisfaction
with the previous one and the source of that dissatisfaction has not previously been discussed in the literature. The difficulty may stem from a poor specification of the keyword assignment task in view of the rank-based approach. With a more complete specification of this task, we aim to show why the previous evaluation metrics fail to satisfy researchers’ goals to distinguish and detect good rank-based keyword assignment systems. We put forward a characterisation of an ideal evaluation metric, and discuss the consistency of the evaluation metrics with this ideal, finding that the average standard normalised cumulative gain metric is most consistent with this ideal.
with the previous one and the source of that dissatisfaction has not previously been discussed in the literature. The difficulty may stem from a poor specification of the keyword assignment task in view of the rank-based approach. With a more complete specification of this task, we aim to show why the previous evaluation metrics fail to satisfy researchers’ goals to distinguish and detect good rank-based keyword assignment systems. We put forward a characterisation of an ideal evaluation metric, and discuss the consistency of the evaluation metrics with this ideal, finding that the average standard normalised cumulative gain metric is most consistent with this ideal.
| Original language | English |
|---|---|
| Title of host publication | Proc of TALN |
| Publisher | Association for Computational Linguistics |
| Publication date | 2015 |
| Publication status | Published - 2015 |
| Externally published | Yes |
Keywords
- Unsupervised Keyword Assignment
- Evaluation Metrics
- Rank-Based Approach
- Normalised Cumulative Gain
- Evaluation Consistency