Social Influence Analysis Research Articles

Social influence analysis and application on multimedia sharing websites

Social media is becoming popular these days, where users necessarily interact with each other to form social networks. Influence network, as one special case of social network, has been recognized as significantly impacting social activities and user decisions. We emphasize in this article that the inter-user influence is essentially topic-sensitive, as for different tasks users tend to trust different influencers and be influenced most by them. While existing research focuses on global influence modeling and applies to text-based networks, this work investigates the problem of topic-sensitive influence modeling in the multimedia domain. According to temporal data justification, we propose a multimodal probabilistic model, considering both users' textual annotation and uploaded visual images. This model is capable of simultaneously extracting user topic distributions and topic-sensitive influence strengths. By identifying the topic-sensitive influencer, we are able to conduct applications, like collective search and collaborative recommendation. A risk minimization-based general framework for personalized image search is further presented, where the image search task is transferred to measure the distance of image and personalized query language models. The framework considers the noisy tag issue and enables easy incorporation of social influence. We have conducted experiments on a large-scale Flickr dataset. Qualitative as well as quantitative evaluation results have validated the effectiveness of the topic-sensitive influencer mining model, and demonstrated the advantage of incorporating topic-sensitive influence in personalized image search and topic-based image recommendation.

Scalable Influence Analysis in Mobile Social Networks

Influence is a complex and subtle force that governs the dynamics of social networks as well as the behaviors of involved users. In large social networks, nodes are influenced by others for various reasons. Understanding influence can benefit various applications such as viral marketing, recommendation, and information retrieval. However, most existing works on social influence analysis have focused on verifying the existence of social influence. In this paper, an algorithm is proposed which utilizes the heterogeneous link information and the textual content associated with each node in the network to mine micro and macro influence. Based on the direct and indirect influence, the MapReduce framework for social influence analysis algorithm is proposed to derive the indirect influence between nodes. We further study how the discovered topic-level influence can help the prediction of user behaviors. Empirical studies on a large real-world mobile social network show that our algorithm has a good scalability performance.

Item-Level Social Influence Prediction with Probabilistic Hybrid Factor Matrix Factorization

Social influence has become the essential factor which drives the dynamic evolution process of social network structure and user behaviors. Previous research often focus on social influence analysis in network-level or topic-level. In this paper, we concentrate on predicting item-level social influence to reveal the users' influences in a more fine-grained level. We formulate the social influence prediction problem as the estimation of a user-post matrix, where each entry in the matrix represents the social influence strength the corresponding user has given the corresponding web post. To deal with the sparsity and complex factor challenges in the research, we model the problem by extending the probabilistic matrix factorization method to incorporate rich prior knowledge on both user dimension and web post dimension, and propose the Probabilistic Hybrid Factor Matrix Factorization (PHF-MF) approach. Intensive experiments are conducted on a real world online social network to demonstrate the advantages and characteristics of the proposed method.

Weight matrices for social influence analysis: An investigation of measurement errors and their effect on model identification and estimation quality

Weight matrices, such as used in network autocorrelation models, are useful to investigate social influence processes. The objective of this paper is to investigate a key topic that has received relatively little attention in previous research, namely the issues that arise when observational limitations lead to measurement errors in these weight matrices. Measurement errors are investigated from two perspectives: when relevant ties are omitted, and when irrelevant ties are erroneously included as part of the matrix. The paper first shows analytically that these two situations result in biased estimates. Next, a simulation experiment provides evidence of the effect of erroneously coding the weight matrix on model performance and the ability of a network autocorrelation test to identify social influence effects. The results suggest that depending on the level of autocorrelation and the topology attributes of the underlying matrix, there is a window of opportunity to identify and model social influence processes even in situations where the ties in a matrix cannot be accurately observed.

Qualitative adolescent health research — focus groups: a rural South African example

This paper introduces nine steps that are recommended in conducting focus group discussions in rural communities and gives an example of how they can appropriately and fruitfully be employed in adolescent health behavioural research. The paper also reviewed issues related to methods of data collection, data analysis, reliability and validity in qualitative research. Focus group discussions took place in classrooms in three schools in Mankweng, Limpopo Province of South Africa. Three groups (boys only, girls only and mixed) took part in each school. Participants were selected from the pool of standard seven (grade 9) students from the chosen schools. The nine steps that were involved in using focus group discussions as a research method and the Mankweng experience is discussed. These steps include: (1) conducting a social influence analysis; (2) identifying the specific information to collect; (3) designing focus group discussion guide; (4) choosing the participants for the focus group discussion; (5) selecting focus group discussion moderators; (6) training focus group discussion moderators; (7) conducting the focus group discussion; (8) analysing the data collected; (9) formulating study conclusions and policy recommendations. Little adolescent health research in South Africa has been based upon methods that can capture the complexity of the role of significant others in adolescent health and development and the powerlessness of rural communities in dealing with the 'new morbidity' of adolescent risk behaviours. Understanding what sort of power relations, for example, that are involved in being relatively disadvantaged and how the power of such social groups can be increased is common concern of development managers and other individuals and institutions engaged in policy changes and implementation and deserve to be an essential component of child and adolescent health research. Well-collected and well-analysed qualitative data is needed in order to clearly understand some of the underlying predisposing, enabling and reinforcing factors that could account for adolescent risk behaviours in rural communities. Qualitative research method is important in that they capture the complexity of life, rather than trying to 'reduce' this complexity, and the potential for gaining an understanding of what is going on in the situation with the phenomenon is greater. Focus group discussions, a qualitative research methodology, can yield valuable data.

Qualitative community health research: a Tunisian example

A group interview method is used to study community perceptions and treatment of diarrheal disease in the Tunisian Diarrheal Disease Control Project (TDDCP). The projects goal is to help mothers manage diarrhea at home. Parents and grandmothers of young children in the 7 cultural regions of Tunisia were studied by a qualitative data collection method. The TDDCP team used the semi-structured interview methodology sometimes called a focus group. 9 steps are involved: 1) a social influence analysis is conducted; 2) specific information to be collected is identified; 3) group interview guides are designed; 4) a sample of individuals to be interviewed are chosen; 5) group interview facilitators are chosen; 6) group interview facilitators are trained; 7) group interviews are done: 8) data is collected and analyzed; and 9) study conclusions are formulated as are policy recommendations for the development of health training and education programs. The study was done during a 4-week August-September period the peak diarrhea season. 61 groups of women were interviewed--642 women an average of 10.5 per group. Interview participants had a low income; were married with at least 1 preschool child--under 4 years of age; already knew each other; and not more than 3 were from the same family. Some courses for diarrhea were scientifically proven; others were metaphysical. Mothers do not perceive dehydration to be a complication of diarrhea. Many continue to breastfeed. The majority give some liquids but no in sufficient quantity. Most mothers continue to feed the child. Many teas/infusions are prepared. 30 different plants are used. These are used to cure the diarrhea. (Plant analysis showed that most contain astringents which have a constipating effect.) Mothers usually try to treat diarrhea at home before they go to a health center. If they perceive the cause to be metaphysical mothers will more frequently consult informal health specialists. Health education planners should not assume that communities are inexperienced or ignorant of dealing with diarrhea. Knowledge and experience do exist. The health education program content should consider both the biomedical guidelines and community practices on a given problem. A program that does not take sociocultural realities into consideration will not be very good. (authors modified)