Sentiment Analysis with Incremental Human-in-the-Loop Learning and Lexical Resource Customization. Mishra, S.; Diesner, J.; Byrne, J.; and Surbeck, E. In Proceedings of the 26th ACM Conference on Hypertext & Social Media, pages 323-325, 2015.
Sentiment Analysis with Incremental Human-in-the-Loop Learning and Lexical Resource Customization [link]Website  abstract   bibtex   
The adjustment of probabilistic models for sentiment analysis to changes in language use and the perception of products can be realized via incremental learning techniques. We provide a free, open and GUI-based sentiment analysis tool that allows for a) relabeling predictions and/or adding labeled instances to retrain the weights of a given model, and b) customizing lexical resources to account for false positives and false negatives in sentiment lexicons. Our results show that incrementally updating a model with information from new and labeled instances can substantially increase accuracy. The provided solution can be particularly helpful for gradually refining or enhancing models in an easily accessible fashion while avoiding a) the costs for training a new model from scratch and b) the deterioration of prediction accuracy over time.
@inProceedings{
 title = {Sentiment Analysis with Incremental Human-in-the-Loop Learning and Lexical Resource Customization},
 type = {inProceedings},
 year = {2015},
 identifiers = {[object Object]},
 keywords = {incremental learning,lexical resource customization,sentiment analysis},
 pages = {323-325},
 websites = {http://doi.acm.org/10.1145/2700171.2791022},
 id = {1870121a-a180-340b-9b83-313697c2a142},
 created = {2016-04-21T16:33:01.000Z},
 file_attached = {false},
 profile_id = {ee28df9d-6898-3d7c-b555-58915f721481},
 last_modified = {2017-03-22T03:35:31.512Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {true},
 hidden = {false},
 citation_key = {Mishra2015},
 abstract = {The adjustment of probabilistic models for sentiment analysis to changes in language use and the perception of products can be realized via incremental learning techniques. We provide a free, open and GUI-based sentiment analysis tool that allows for a) relabeling predictions and/or adding labeled instances to retrain the weights of a given model, and b) customizing lexical resources to account for false positives and false negatives in sentiment lexicons. Our results show that incrementally updating a model with information from new and labeled instances can substantially increase accuracy. The provided solution can be particularly helpful for gradually refining or enhancing models in an easily accessible fashion while avoiding a) the costs for training a new model from scratch and b) the deterioration of prediction accuracy over time.},
 bibtype = {inProceedings},
 author = {Mishra, Shubhanshu and Diesner, Jana and Byrne, Jason and Surbeck, Elizabeth},
 booktitle = {Proceedings of the 26th ACM Conference on Hypertext & Social Media}
}
Downloads: 0