doi: 10.18178/ijfcc.2015.4.6.419
Predicting Latent Trends of Labels in the Social Media Using Infectious Capacity
Yuki Sonoda
,
Ikeda Daisuke
- Department of Informatics, Kyushu University, Japan.
Abstract
This paper is devoted to predicting trends on thesocial media. Typical methods in the literature are based ontemporal changes in usage of words or phrases on the media,and try to find a rapid increase, called a burst, of them.Therefore, these methods can be applied only after a burst isemerging. In this paper, we propose an index, called theinfectious capacity, to detect potential trends on the social mediabefore they would emerge. To achieve this, we focus on labelsand items, and predict trends of a label, instead of those of atarget object, such as contents of a social media, where an itemis a concept represented by an object and a label categoriesitems. On a photo sharing service, for example, a photo is anobject, a tag is a label, and concepts represented by a photo areitems for the photo. Using labels and items, the infectiouscapacity for a label is defined as the ratio of the variety of itemswith the label to the number of occurrences of the label in givendata. That is, the larger value an infectious capacity of a label is,more infectious the label is. Our experiments on real datashowed that the infectious capacities for most labels aresubstantially constant over time. This result means that we canforecast the variety per usage for a label just after the label isused. Moreover, we found that infectious capacities for popularlabels have similar values. Combined with the first result, weare able to predict latent trends before labels become popular.In fact, this is also supported by experiments on tweets, wherewe were able to find potentially popular hashtags, regardinghashtags as labels, before they become popular. As far as theauthors know, this is the first result of future trend predictionon the social media.
Keywords
- Category
- constant index
- future trenddetection
How to Cite
Yuki Sonoda and Ikeda Daisuke, "Predicting Latent Trends of Labels in the Social Media Using Infectious Capacity," International Journal of Future Computer and Communication, vol. 4, no. 6, pp. 374-380, 2015. https://doi.org/ 10.18178/ijfcc.2015.4.6.419
Copyright & License
Copyright © 2015 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).