Mobile App Classification with Enriched Contextual Information

Dong

2014 IEEE International Conference on Data Mining

et al. 2014

Self Cite

The emergence of social networks has provided opportunities for both targeted marketing and viral marketing. By concentrating the efforts on a few key customers, targeted marketing could make the promotion of the items (products) much easier and more cost-effective. On the other hand, viral marketing aims at ¿nding a set of individuals (seeds) to maximize the word-of-mouth propagation of an item. However, these two marketing strategies can only exploit some speci¿c characteristics of the social networks, and the problem of how to combine them together to build a better, stronger business is still open. To that end, in this paper, we propose a general approach for integrated marketing. Speci¿cally, to market a given item, we ¿rst generate the item-speci¿c candidate users by a recommendation algorithm, and then select the typical users who have the best balanced utility scores and consumption/social entropy. Next, treating typical users as targeted customers, we study the problem of maximizing information awareness in viral marketing with these constrained targets. Along this line, we de¿ne it as a constrained coverage maximization problem, and propose three solutions: GMIC, LMIC and QMIC. Finally, extensive experimental results on real-world datasets demonstrate that our integrated marketing approach could outperform the methods that consider only targeted marketing or viral marketing.

Section: Discussionmentioning

confidence: 99%

Social Marketing Meets Targeted Customers: A Typical User Selection and Coverage Perspective

Dong

2014 IEEE International Conference on Data Mining

et al. 2014

Self Cite

“…Classification is a very fundamental problem in machine learning [26,39,40]. Traditionally, sentiment classification is often addressed by classic classification approaches directly.…”

Section: Sentiment Analysismentioning

confidence: 99%

Tracking the evolution of social emotions with topic models

Zhu

et al. 2015

Knowl Inf Syst

Self Cite

Many of today's online news Web sites have enabled users to specify different types of emotions (e.g., angry or shocked) they have after reading news. Compared with traditional user feedbacks such as comments and ratings, these specific emotion annotations are more accurate for expressing users' personal emotions. In this paper, we propose to exploit these users' emotion annotations for online news in order to track the evolution of emotions, which plays an important role in various online services. A critical challenge is how to model emotions with respect to time spans. To this end, we propose a timeaware topic modeling perspective for solving this problem. Specifically, we first develop two models named emotion-Topic over Time (eToT) and mixed emotion-Topic over Time (meToT), in which the topics of news are represented as a beta distribution over time and a multinomial distribution over emotions. While they can uncover the latent relationship among news, emotion and time directly, they cannot capture the evolution of topics. Therefore, we further develop another model named emotion-based Dynamic Topic Model (eDTM), where we explore the state space model for tracking the evolution of topics. In addition, we demonstrate that all of proposed models could enable several potential applications, such as emotion prediction, emotion-based news recommendations, and emotion anomaly detections. Finally, we validate the proposed models with extensive experiments with a real-world data set.

2014 IEEE International Conference on Data Mining

“…Classification is a very fundamental problem in machine learning [25], [33], [34]. Traditionally, sentiment classification is often formalized as a classification task, and thus could be addressed by the classification approaches directly.…”

Section: A Sentiment Analysismentioning

confidence: 99%

Tracking the Evolution of Social Emotions: A Time-Aware Topic Modeling Perspective

Zhu

et al. 2014

Self Cite

Many of today's online news websites have enabled users to specify different types of emotions (e.g., angry and shocked) they have after reading news. Compared with traditional user feedbacks such as comments and ratings, these specific emotion annotations are more accurate for expressing users' personal emotions. In this paper, we propose to exploit these users' emotion annotations for online news in order to track the evolution of emotions, which plays an important role in various online services. A critical challenge is how to model emotions with respect to time spans. To this end, we propose a time-aware topic modeling perspective for solving this problem. Specifically, we first develop a model named emotion-Topic over Time (eToT), in which we represent the topics of news as a Beta distribution over time and a multinomial distribution over emotions. While eToT can uncover the latent relationship among news, emotion and time directly, it cannot capture the dynamics of topics. Therefore, we further develop another model named emotionbased Dynamic Topic Model (eDTM), where we explore the state space model for tracking the dynamics of topics. In addition, we demonstrate that both eToT and eDTM could enable several potential applications, such as emotion prediction, emotion-based news recommendations and emotion anomaly detections. Finally, we validate the proposed models with extensive experiments with a real-world data set.