Dataset Information

Predicting Positive and Negative Relationships in Large Social Networks.

ABSTRACT: In a social network, users hold and express positive and negative attitudes (e.g. support/opposition) towards other users. Those attitudes exhibit some kind of binary relationships among the users, which play an important role in social network analysis. However, some of those binary relationships are likely to be latent as the scale of social network increases. The essence of predicting latent binary relationships have recently began to draw researchers' attention. In this paper, we propose a machine learning algorithm for predicting positive and negative relationships in social networks inspired by structural balance theory and social status theory. More specifically, we show that when two users in the network have fewer common neighbors, the prediction accuracy of the relationship between them deteriorates. Accordingly, in the training phase, we propose a segment-based training framework to divide the training data into two subsets according to the number of common neighbors between users, and build a prediction model for each subset based on support vector machine (SVM). Moreover, to deal with large-scale social network data, we employ a sampling strategy that selects small amount of training data while maintaining high accuracy of prediction. We compare our algorithm with traditional algorithms and adaptive boosting of them. Experimental results of typical data sets show that our algorithm can deal with large social networks and consistently outperforms other methods.

SUBMITTER: Wang GN

PROVIDER: S-EPMC4468140 | biostudies-literature | 2015

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting Positive and Negative Relationships in Large Social Networks.

Wang Guan-Nan GN Gao Hui H Chen Lian L Mensah Dennis N A DN Fu Yan Y

PloS one 20150615 6

In a social network, users hold and express positive and negative attitudes (e.g. support/opposition) towards other users. Those attitudes exhibit some kind of binary relationships among the users, which play an important role in social network analysis. However, some of those binary relationships are likely to be latent as the scale of social network increases. The essence of predicting latent binary relationships have recently began to draw researchers' attention. In this paper, we propose a m ...[more]

PMID: 26075404

Similar Datasets

Project description:BackgroundWhile online health social networks (OHSNs) serve as an effective platform for patients to fulfill their various social support needs, predicting the needs of users and providing tailored information remains a challenge.ObjectiveThe objective of this study was to discriminate important features for identifying users' social support needs based on knowledge gathered from survey data. This study also provides guidelines for a technical framework, which can be used to predict users' social support needs based on raw data collected from OHSNs.MethodsWe initially conducted a Web-based survey with 184 OHSN users. From this survey data, we extracted 34 features based on 5 categories: (1) demographics, (2) reading behavior, (3) posting behavior, (4) perceived roles in OHSNs, and (5) values sought in OHSNs. Features from the first 4 categories were used as variables for binary classification. For the prediction outcomes, we used features from the last category: the needs for emotional support, experience-based information, unconventional information, and medical facts. We compared 5 binary classifier algorithms: gradient boosting tree, random forest, decision tree, support vector machines, and logistic regression. We then calculated the scores of the area under the receiver operating characteristic (ROC) curve (AUC) to understand the comparative effectiveness of the used features.ResultsThe best performance was AUC scores of 0.89 for predicting users seeking emotional support, 0.86 for experience-based information, 0.80 for unconventional information, and 0.83 for medical facts. With the gradient boosting tree as our best performing model, we analyzed the strength of individual features in predicting one's social support need. Among other discoveries, we found that users seeking emotional support tend to post more in OHSNs compared with others.ConclusionsWe developed an initial framework for automatically predicting social support needs in OHSNs using survey data. Future work should involve nonsurvey data to evaluate the feasibility of the framework. Our study contributes to providing personalized social support in OHSNs.

Project description:Note: Data Use Restrictions: <ul> <li>The data must only be used for genetic research projects solely in the area of health-related social-network research.</li> </ul> This study compiles detailed information on family and social ties linked to participants in the offspring cohort of the Framingham Heart Study. The investigators took computerized information from self-report data used by the FHS for over 30 years to facilitate health exam scheduling. These tracking sheets asked participants to identify people close to them, which were assumed by the investigators to indicate a social tie. Individuals are listed as "egos" (if the ties are from their perspective) and/or "alters" (if they are linked as a friend or family members to one of the other participants). All egos were FHS offspring participants while alters could be FHS participants in any cohort or non-participants. This information was combined with internal FHS pedigree data on family ties to list family member alters (as well as the nature of their relationship). The resultant dataset therefore includes each ego-alter tie, the nature of their relationship and the start and end dates for their ties for each of seven waves for the offspring study. Due to privacy concerns, all exam dates for individuals are listed relative to the initial FHS exam date. The study investigators chose a random, confidential date, to which all initial exam dates are linked (by number of months from random date to initial exam date). With this information, researchers can have access to the temporal relationships among participants' exam dates while remaining unaware of the actual exam dates for individual participants. During each clinic exam cycle, the participants undergo a detailed examination including physical examination, medical history, laboratory testing, and electrocardiogram. Over the years, other tests (that may not be performed at every exam cycle) have included pulmonary function, lifestyle, physical function, cognitive function questionnaires, and various noninvasive cardiovascular tests including echocardiograms. This study involves phenotypic data from these exams including (when available) basic body measurements (height, weight, blood pressure), laboratory values (blood sugar levels, LDL levels), information on smoking and alcohol use, and tests of depression and cognitive functioning. Important links to apply for individual-level data <ol> <li><a href="http://dbgap.ncbi.nlm.nih.gov/aa/wga.cgi?view_pdf&stacc=phs000153.v3.p3" target="_blank">Data Use Certification Requirements (DUC)</a></li> <li><a href="GetPdf.cgi?id=phd001183.1" target="_blank">Instructions to Request Authorized Access</a></li> <li><a href="http://view.ncbi.nlm.nih.gov/dbgap-controlled" target="_blank">Apply here for controlled access to individual level data</a></li> <li><a href="GetPdf.cgi?id=phd000317" target="_blank">Participant Protection Policy FAQ</a></li> </ol> In 1948, researchers recruited men and women from the town of Framingham, Massachusetts, beginning the first round of extensive physical examinations and lifestyle interviews that would later be analyzed for common patterns related to CVD development. Initially, the Framingham Heart Study enrolled 5,209 men and women from the Framingham area who were between the ages of 28 and 62 years. Beginning in 1971, the Framingham Heart Study enrolled 5,124 men and women, who were either offspring of the original cohort or spouses of those offspring. In 2002, 4,095 third generation participants (men and women) were enrolled.

Project description:IntroductionEngaging in teamwork requires a clear understanding of positive and negative behaviours that act as facilitators and barriers to collegial workplace relationships. Identifying and correcting underlying barriers, while promoting facilitators, is fundamental to improving care delivery and, ultimately, clinical outcomes. Despite a considerable amount of literature in this area, there is a lack of clarity of the different behaviours as several parallel literatures address similar questions about antecedents, processes and outcomes. The purpose of this study is to synthesise the current state of literature reporting on behaviours in workplace relationships. Using a scoping review methodology, the following research question will be addressed: "What is known about positive and negative behaviours in workplace relationships?"Methods and analysisWe will employ the methodological frameworks used by Arksey and O'Malley and Levac et al. The search strategy will include numerous electronic databases, grey literature sources and hand-searching of reference lists from 1990 to present with a limit to English language. Search strategies will be developed using controlled vocabulary and keyword terms related to various components of workplace relationships. Two reviewers will independently screen titles and abstracts for inclusion, followed by screening of the full text of potential articles to determine final inclusion. A descriptive numerical analysis will describe characteristics of included studies. A thematic analysis will provide an overview of the literature, including definitions, conceptual frameworks, antecedents, outcomes and interventions.DisseminationIn reviewing a wide range of positive and negative behaviours, then integrating into a manageable, meaningful whole, this study is a critical step in helping policymakers, leaders and healthcare professionals effectively use what is known thus far. Knowledge translation activities will occur throughout the study with dissemination of findings to local, national, and international stakeholders, including a wide range of clinicians, leaders and administrators in all sectors.

Dataset Information

Predicting Positive and Negative Relationships in Large Social Networks.

Publications

Predicting Positive and Negative Relationships in Large Social Networks.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets