User-generated reviews are valuable resources for consumers to gain information of products which has significant impact on their following decision-making. With the development of social network service, consumers are exposed to reviews coming from both friends and the crowds (non-friends). However, the impact of friends' and crowds' reviews on consumer posting behaviour has not been well differentiated. Using the online review information as well as the underlying social network from Yelp, this paper develops a multilevel mixed effect probit model to study the impact of consumer characteristics and reviews of different sources, i.e. friends or crowds, on the possibility of consumer further engaging in posting behaviour. Despite the common perception that the volume, valance and variance of reviews significantly impact the possibility of following posting behaviour, we show that such influence majorly comes from the friend reviews. The volume of friend reviews has much stronger impact on the target user's posting behaviour than that of the crowds. The valance and variance of the crowd reviews show no significant influence when ignoring the friend reviews, but negative influence when considering it. The friend reviews and crowd reviews are further divided as positive and negative ones, and only the positive friend reviews and negative crowd review are found significantly enhancing the posting possibility.
The mechanism of the online user preference evolution is of great significance for understanding the online user behaviors and improving the quality of online services. Since users are allowed to rate on objects in many online systems, ratings can well reflect the users' preference. With two benchmark datasets from online systems, we uncover the memory effect in users' selecting behavior which is the sequence of qualities of selected objects and the rating behavior which is the sequence of ratings delivered by each user. Furthermore, the memory duration is presented to describe the length of a memory, which exhibits the power-law distribution, i.e., the probability of the occurring of long-duration memory is much higher than that of the random case which follows the exponential distribution. We present a preference model in which a Markovian process is utilized to describe the users' selecting behavior, and the rating behavior depends on the selecting behavior. With only one parameter for each of the user's selecting and rating behavior, the preference model could regenerate any duration distribution ranging from the power-law form (strong memory) to the exponential form (weak memory).
Similarity is a fundamental measure in network analyses and machine learning algorithms, with wide applications ranging from personalized recommendation to socio-economic dynamics. We argue that an effective similarity measurement should guarantee the stability even under some information loss. With six bipartite networks, we investigate the stabilities of fifteen similarity measurements by comparing the similarity matrixes of two data samples which are randomly divided from original data sets. Results show that, the fifteen measurements can be well classified into three clusters according to their stabilities, and measurements in the same cluster have similar mathematical definitions. In addition, we develop a top-n-stability method for personalized recommendation, and find that the unstable similarities would recommend false information to users, and the performance of recommendation would be largely improved by using stable similarity measurements. This work provides a novel dimension to analyze and evaluate similarity measurements, which can further find applications in link prediction, personalized recommendation, clustering algorithms, community detection and so on.
Despite enormous recent efforts in detecting the mechanism of the social relation formation in online social systems, the underlying rules between the common interests and social relations are still under dispute. Do online users befriend others who have similar tastes, or do their tastes become more similar after they become friends? In this paper, we investigate the correlation between online user trust formation and their common interests, measured by the overlap rate ρ and taste similarity θ respectively. The trust relation creation time is set as the zero timestamp. The statistical results before and after the trust formation for an online network, namely Epinions, show that, the overlap rate ρ increases greatly before the trust formation, while it would increase smoothly after the creation of the trust relation. Comparing with the empirical results, two null models are presented by shuffling the temporal behaviors of online users, which suggests that the accumulation of the common interests can result in the trust formation. Furthermore, we investigate the taste similarity θ of the common interests, which can reflect the users’ preference on their common interests. The empirical results show that the taste similarity θ is rapidly increased around the day when users trust the others. That is, the similar tastes on the common interests among users lead to the trust formation. Finally, we report that the user degree can also influence the effect of the taste similarity θ on user trust formation. This work may shed some light for deeply understanding the evolution mechanism of the online social systems.
The structures of feature vectors based semi-supervised/supervised learning has gained considerable interests in the past several years thanks to its effectiveness for better object modeling and classification. In many machine learning and computer vision tasks, a critical issue is the similarity between two feature vectors. In this paper, we present a novel technique to measure the similarities among feature vectors by decomposing each feature vector as an 1 sparse linear combination of the rest of the feature vectors. The main idea is that the coefficients in such a sparse decomposition reflect the features' neighborhood structure thus providing better similarity measures among the decomposed feature vector and the rest of the feature vectors. The proposed approach is applied to label propagation and action recognition, and is evaluated on several commonly-used data sets. The experimental results show that the proposed Sparsity Induced Similarity (SIS) measure significantly improves the performance of both label propagation and action recognition.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.