EAI Endorsed Transactions on Scalable Information Systems 15(6): e5

Research Article

360-MAM-Affect: Sentiment Analysis with the Google Prediction API and EmoSenticNet

Download96 downloads
  • @ARTICLE{10.4108/icst.intetain.2015.259631,
        author={Eleanor Mulholland and Paul Mc Kevitt and Tom Lunney and John Farren and Judy Wilson},
        title={360-MAM-Affect: Sentiment Analysis with the Google Prediction API and EmoSenticNet},
        journal={EAI Endorsed Transactions on Scalable Information Systems},
        volume={15},
        number={6},
        publisher={EAI},
        journal_a={SIS},
        year={2015},
        month={8},
        keywords={affective computing, emosenticnet, gamification, google prediction api, head squeeze, machine learning, natural language processing, recommender system, sentiment analysis, youtube, 360-mam-affect, 360-mam-select},
        doi={10.4108/icst.intetain.2015.259631}
    }
    
  • Eleanor Mulholland
    Paul Mc Kevitt
    Tom Lunney
    John Farren
    Judy Wilson
    Year: 2015
    360-MAM-Affect: Sentiment Analysis with the Google Prediction API and EmoSenticNet
    SIS
    EAI
    DOI: 10.4108/icst.intetain.2015.259631
Eleanor Mulholland1,*, Paul Mc Kevitt1, Tom Lunney1, John Farren2, Judy Wilson2
  • 1: Ulster University, School of Creative Arts & Technologies
  • 2: 360 Production Ltd.
*Contact email: mulholland-e9@email.ulster.ac.uk

Abstract

Online recommender systems are useful for media asset management where they select the best content from a set of media assets. We have developed an architecture for 360-MAM- Select, a recommender system for educational video content. 360-MAM-Select will utilise sentiment analysis and gamification techniques for the recommendation of media assets. 360-MAM-Select will increase user participation with digital content through improved video recommendations. Here, we discuss the architecture of 360-MAM-Select and the use of the Google Prediction API and EmoSenticNet for 360-MAM-Affect, 360-MAM-Select's sentiment analysis module. Results from testing two models for sentiment analysis, Sentiment Classifier (Google Prediction API) and EmoSenticNetClassifer (Google Prediction API + EmoSenticNet) are promising. Future work includes the implementation and testing of 360-MAM-Select on video data from YouTube EDU and Head Squeeze.