Search

In traditional group decision-making models, it is commonly assumed that all decision makers exert equal influence on one another. However, in real-world social networks, such as Twitter and Facebook, certain individuals—known as top persuaders—hold a disproportionately large influence over others. This study formulates the consensus-reaching problem in social network group decision making by introducing a novel framework for predicting top persuaders. Building on social network theories, we develop a social persuasion model that integrates social influence and social status to quantify individuals' persuasive power more comprehensively. Subsequently, we propose a new CRP that leverages the influence of top persuaders. Our simulations and comparative analyses demonstrate that: (1) increasing the number of top persuaders substantially reduces the iterations required to achieve consensus; (2) establishing trust relationships between top persuaders and other individuals accelerates the consensus process; and (3) top persuaders retain a high and stable level of influence throughout the entire CRP rounds. Our research provides practical insights into identifying and strategically guiding top persuaders to enhance the efficiency in consensus reaching and reduce social management costs within social networked environments. [ABSTRACT FROM AUTHOR]

Online and offline integration is an increasingly popular method of performing modern medical services. To provide suggestions for physician selection by patients with full use of previous online and offline patient evaluations, this paper investigates online and offline social network group decision making (OAOSNGDM) in depth. With the aim of inferring the indirect trust relationships of online and offline patients, we first construct a q-rung orthopair fuzzy dual trust propagation operator based on the q-rung orthopair fuzzy trust function, which can effectively deal with inconsistency in trust functions among patients. Considering patient inconsistency in online and offline scenarios, which can increase the uncertainty of the trust relationship in cross-network propagation, we propose a q-rung orthopair fuzzy dual trust cross-network propagation operator by introducing cross-network propagation efficiency. Considering the signal-to-noise ratio, we calculate the trust propagation efficiency and introduce it into the trust propagation operators. To aggregate the trust information of multiple trust paths among patients, we introduce the Dempster rule from evidence theory which can handle the uncertainty of trust functions. In addition, to accurately determine the patient weights according to online and offline social networks, we integrate the ranking results of patients in terms of degree centrality, neighbor importance and betweenness centrality by developing an improved linear assignment method. We then propose a novel decision-making method for OAOSNGDM and design a complete decision-making process for the evaluation of physicians. Finally, we verify the effectiveness of our proposed method for the evaluation of physicians in an online and offline scenario. [ABSTRACT FROM AUTHOR]

The development of the Internet of Things (IoT) has brought great convenience to people. However, some information security problems such as privacy leakage are caused by communicating with risky users. It is a challenge to choose reliable users with which to interact in the IoT. Therefore, trust plays a crucial role in the IoT because trust may avoid some risks. Agents usually choose reliable users with high trust to maximize their own interests based on reinforcement learning. However, trust propagation is time-consuming, and trust changes with the interaction process in social networks. To track the dynamic changes in trust values, a dynamic trust inference algorithm named Dynamic Double DQN Trust (Dy-DDQNTrust) is proposed to predict the indirect trust values of two users without direct contact with each other. The proposed algorithm simulates the interactions among users by double DQN. Firstly, CurrentNet and TargetNet networks are used to select users for interaction. The users with high trust are chosen to interact in future iterations. Secondly, the trust value is updated dynamically until a reliable trust path is found according to the result of the interaction. Finally, the trust value between indirect users is inferred by aggregating the opinions from multiple users through a Modified Collaborative Filtering Average-based Similarity (SMCFAvg) aggregation strategy. Experiments are carried out on the FilmTrust and the Epinions datasets. Compared with TidalTrust, MoleTrust, DDQNTrust, DyTrust and Dynamic Weighted Heuristic trust path Search algorithm (DWHS), our dynamic trust inference algorithm has higher prediction accuracy and better scalability.

Collaborative filtering (CF) is the most popular approach for predicting relevant items in recommender systems. However, basic CF suffers from some serious problems including data sparsity and cold start. Incorporating trust inferences into traditional collaborative filtering is an effective approach to overcoming these problems and obtaining more accurate recommendations. Since the value of direct trust in a user is not always available, trust propagation, i.e., indirect estimation of the trust level, may be helpful in a trust-based recommendation. The effectiveness and accuracy of trust-based recommendation systems may be improved if different parameters are taken into account in trust propagation. In this paper, we introduce a four-dimensional trust propagation model for use in recommendation systems in which social distance, location, time, and context are all taken into account. In the proposed model, firstly using the subjective logic model, a confidence-aware trust propagation procedure is run to estimate the indirect trust values based on social links between users. Then, a compound similarity measure is calculated based on the closeness among ratings in terms of time, location, and context. This similarity measure is used to estimate final trust values among users. Finally, rating estimation is done using the levels of trust as weights. The results of the conducted experiments on well-known datasets demonstrate that the proposed model provides higher effectiveness and accuracy comparing the existing methods. An ablation analysis was conducted to evaluate the contribution of each feature dimension to the proposed model. Also, the complexity of the method was analyzed, which confirms the scalability of the model. [ABSTRACT FROM AUTHOR]

The existing online medical platforms are striving to provide various kinds of services to satisfy users' dynamic demands. With the exponential growth of information, users usually suffer from information overload, confronting with high search costs and time costs. In order to reduce users' search cost and assist users to retrieve appropriate medical service efficiently, recommending online medical services to users smarter and faster has becoming a valuable research topic. Unfortunately, certain imprecise and subjective information always exist in obtaining users' preferences. Moreover, the effect of social trust on an individual user's decision-making is often omitted in previous methods, which may result in inaccurate recommendation results. This study proposes a method for recommending the most appropriate medical services to users, in which users' linguistic preferences, assessments and social trust information are considered. The procedure of the proposed method consists of two steps: (1) to acquire various users' linguistic preference based on Best–Worst Method (BWM) and fuzzy set theory, and (2) to recommend the corresponding medical services using trust-aware collaborative filtering (CF) technique. Firstly, the fuzzy BWM is adopted to determine the attribute weights based on various users' linguistic preferences. Secondly, by combing the linguistic evaluation matrix and users' social trust information, the trust-based similarity network is constructed, which is then merged into CF technique for predicting final rating and recommending online medical services. Finally, a case study is conducted to demonstrate the applications of the proposed method. [ABSTRACT FROM AUTHOR]

The growing concern about the risk and safety of autonomous vehicles (AVs) has made it vital to understand driver trust and behavior when operating AVs. While research has uncovered human factors and design issues based on individual driver performance, there remains a lack of insight into how trust in automation evolves in groups of people who face risk and uncertainty while traveling in AVs. To this end, we conducted a naturalistic experiment with groups of participants who were encouraged to engage in conversation while riding a Tesla Model X on campus roads. Our methodology was uniquely suited to uncover these issues through naturalistic interaction by groups in the face of a risky driving context. Conversations were analyzed, revealing several themes pertaining to trust in automation: (1) collective risk perception, (2) experimenting with automation, (3) group sense-making, (4) human-automation interaction issues, and (5) benefits of automation. Our findings highlight the untested and experimental nature of AVs and confirm serious concerns about the safety and readiness of this technology for on-road use. The process of determining appropriate trust and reliance in AVs will therefore be essential for drivers and passengers to ensure the safe use of this experimental and continuously changing technology. Revealing insights into social group–vehicle interaction, our results speak to the potential dangers and ethical challenges with AVs as well as provide theoretical insights on group trust processes with advanced technology. [ABSTRACT FROM AUTHOR]

• We propose a concept of two-tuple linguistic trust/distrust relationship (LTR). • The propagation operator of indirect two-tuple LTR is presented. • A three-level consensus identification rule which considers self-confidence is proposed. • Adjustment rule which considers trust relationship, consensus degree and reliability is analyzed. Large-scale group decision making (LSGDM) is very common in real world, and especially how to reach a relatively consensus status in a social network is a hot topic. In this paper, we propose a concept of two-tuple linguistic trust/distrust relationship (LTR) which could present both trust and distrust degrees by semantics. The trust/distrust representation scheme can unburden individuals from providing numerical trust or distrust degree to just presenting linguistic variables. Transformation rule from two-tuple LTR to numerical trust degree is then analyzed, followed by the propagation operator of indirect trust/distrust relationships. The advantage of the propagation operator lies in that ignorance in trust/distrust relationships can be tackled rationally. As for the consensus reaching process (CRP), three-level identification and adjustment mechanisms are proposed under the condition that individuals express their preferences in an uncertain distribution form. Trust relationship, consensus degree and reliability of individuals' judgments are all addressed comprehensively to narrow the opinion divergence. Self-confidence extent is utilized as a factor to adjust the opinions of non-consensus experts. The proposed method is further implemented in life cycle sustainability assessment to demonstrate the validity and effectiveness in dealing with realistic GDM problems. [ABSTRACT FROM AUTHOR]

Computing the number of k-hop paths is crucial for selecting services in social networks and analyzing graph data, for example, a service consumer require to evaluate the trustworthiness of a service provider along the social trust paths from a service consumer to the service provider, there are usually many social trust paths between two unconnected participants, people need to know the number of loop-free k-hop trust propogation paths; other applications include the similarity computation for services recommendation, information diffusion, etc. Previously, the number of k-hop paths is roughly estimated by the elements in the k multiplications of the network adjacency matrix. This method calculates much more k-hop paths than those actually exist, due to many paths with loops counted as k-hop paths, which may result in obvious errors in applications. Based on the idea of loops removing, accurate mathematical formulas for counting loop-free paths are obtained in this article for paths with five or less hops, an approximate method is provided for larger hops. Based on the proposed loop removing algorithm (LRA), the typical method for predicting trust between any two people in social networks is improved, the error rate is dramatically reduced; the traditional path based similarity indices are improved, which are much accurate than their antecedent counterparts; and a method for computing the spreading probability for information spreading between two unconnected vertices in the famous independent cascade (IC) model is also obtained. To reveal the effectiveness of the proposed LRA, this article also provide a traversal depth-first search algorithm (DFSA) for finding the true number of k-hop loop-free paths. [ABSTRACT FROM AUTHOR]

Online social networks are vital environments for information sharing and user interactivity. To help users of online social services to build, expand, and maintain their friend networks or webs of trust, trust management systems have been deployed and trust inference (or more generally, friend recommendation) techniques have been studied in many online social networks. However, there are some challenging issues obstructing the real-world trust inference tasks. Using only explicit yet sparse trust relationships to predict user trust is inefficient in large online social networks. In the age of privacy-respecting Internet, certain types of user data may be unavailable, and thus existing models for trust inference may be less accurate or even defunct. Although some less interpretable models may achieve better performance in trust prediction, the interpretability of the models may prevent them from being adopted or improved for making relevant informed decisions. To tackle these problems, we propose a probabilistic graphical model for trust inference in online social networks in this paper. The proposed model is built upon the skeleton of explicit trust relationships (the web of trust) and embeds various types of available user data as comprehensively-designed trust-aware features. A message passing algorithm, loop belief propagation, is applied to the model inference, which greatly improves the interpretability of the proposed model. The performance of the proposed model is demonstrated by experiments on a real-world online social network dataset. Experimental results show the proposed model achieves acceptable accuracy with both fully and partially available data. Comparison experiments were conducted, and the results show the proposed model's promise for trust inference in some circumstances. [ABSTRACT FROM AUTHOR]

Context-aware Recommender Systems (CARS) deal with modeling and prediction of user interests and preferences according to contextual information while generating a recommendation. In contextual modeling-based CARS, the context information is used straight into the recommendation function as a predictor explicitly. Thus, this approach formulates a multidimensional recommendation model and is best realized through Tensor Factorization (TF) based techniques. It efficiently handles the data sparsity problem faced by most of the traditional RS. However, the recent TF-based CARS face issues such as differentiating amongst relevant and irrelevant context variables, biased recommendations, and long-tail problem. In this paper, we propose a fusion-based approach for determining the list of most relevant and optimum contexts for two datasets, namely the LDos Comoda and Travel dataset. The mutual trust model that combines user level and item level trust is proposed further which utilizes the concept of trust propagation to calculate the inferred trust between users/items. Finally, a hybrid reranking technique combining the item popularity and item absolute likeability reranking approaches with the standard ranking technique of generating recommendations is proposed to generate diversified recommendations. Comparative experiments on the LDos Comoda and the Travel datasets are conducted and the experimental results show an improvement of the proposed work with respect to RMSE of 50%, 55%, and 59% compared to MF-based RS, trust-based RS, and context-aware RS respectively. Also, the proposed reranking technique shows approximately three times more diversified recommendations than the standard ranking approach without a significant loss in precision. [ABSTRACT FROM AUTHOR]

The proper selection of a renewable energy project (REP) is the key to give play to regional resource advantages and optimize the energy consumption structure. As a complex multi-criteria group decision-making (MCGDM) problem, the choice of REP involves many influencing factors such as economy, society, and environment. To solve this problem, this paper proposes a social network MCGDM framework based on social network analysis (SNA) and hesitant probabilistic fuzzy set (HPFS). Firstly, the HPFS is introduced into the social network and constructs the hesitant probabilistic fuzzy trust function to represent the trust relationship between decision-makers. Secondly, the trust propagation operator of the hesitant probabilistic fuzzy trust function is constructed using the idea of conditional probability; then a dual feedback mechanism including opinion modification and weight adjustment is proposed to improve the group consensus level. On this basis, a social network MCGDM framework for REP selection is further proposed, and the SNA method is applied to the selection of REP for the first time. Following this, a case in Zhejiang Province, China, is presented to prove the rationality and effectiveness of the framework. The results show that the proposed hesitant probabilistic fuzzy trust function can better reflect the trust relationship between decision-makers and better describe the process of trust transmission, which is of great significance for solving the MCGDM problems such as REP selection. [ABSTRACT FROM AUTHOR]

With the development of online social networking applications, microblogs have become a necessary online communication network in daily life. Users are interested in obtaining personalized recommendations related to their tastes and needs. In some microblog systems, tags are not available, or the use of tags is rare. In addition, user-specified social relations are extremely rare. Hence, sparsity is a problem in microblog systems. To address this problem, we propose a new framework called Pblog to alleviate sparsity. Pblog identifies users' interests via their microblogs and social relations and computes implicit similarity among users using a new algorithm. The experimental results indicated that the use of this algorithm can improve the results. In online social networks, such as Twitter, the number of microblogs in the system is high, and it is constantly increasing. Therefore, providing personalized recommendations to target users requires considerable time. To address this problem, the Pblog framework groups similar users using the analytic hierarchy process (AHP) method. Then, Pblog prunes microblogs of the target user group and recommends microblogs with higher ratings to the target user. In the experimental results section, the Pblog framework was compared with several other frameworks. All of these frameworks were run on two datasets: Twitter and Tumblr. Based on the results of these comparisons, the Pblog framework provides more appropriate recommendations to the target user than previous frameworks. [ABSTRACT FROM AUTHOR]

It is generally accepted that trust is one of the main issues in the area of social networks. Trust values can be measured explicitly or implicitly. There are various approaches to compute the trust value of users implicitly. These approaches consider a uniform trusting behavior for all users of a social network. While it would be more accurate to acknowledge the differences between the users in terms of how they trust others and the factors that they consider in this regard. Trusting behavior of users can also be influenced by the behavior of others with whom they interact. The mechanism of these influences and the conditions for changing the trusting behavior are of great importance for this discussion. In this study, our goal is to model the differences between the trusting behaviors of social network users. For this purpose, we define three behavior modes for the way users trust each other. In each mode, trust calculations are based on behavioral and functional characteristics of users, which are shaped by their subjective beliefs. A dataset that has the interaction information between each pair of nodes is used to define the trusting behavior pattern between users. Also, three scenarios are defined for assessing the propagation of the behavior modes. Then, we attempt to maximize the influence and find the effective nodes for propagating the trusting behavior modes in the network. To this end, we focus on the structure of the social network users in different propagation scenarios. The results show differences between the trust level outcomes reached with each behavior mode. Also, the impacts of propagation source node selection methods differ in each propagation scenario. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Online social networks are vital environments for information sharing and user interactivity. To help users of online social services to build, expand, and maintain their friend networks or webs of trust, trust management systems have been deployed and trust inference (or more generally, friend recommendation) techniques have been studied in many online social networks. However, there are some challenging issues obstructing the real-world trust inference tasks. Using only explicit yet sparse trust relationships to predict user trust is inefficient in large online social networks. In the age of privacy-respecting Internet, certain types of user data may be unavailable, and thus existing models for trust inference may be less accurate or even defunct. Although some less interpretable models may achieve better performance in trust prediction, the interpretability of the models may prevent them from being adopted or improved for making relevant informed decisions. To tackle these problems, we propose a probabilistic graphical model for trust inference in online social networks in this paper. The proposed model is built upon the skeleton of explicit trust relationships (the web of trust) and embeds various types of available user data as comprehensively-designed trust-aware features. A message passing algorithm, loop belief propagation, is applied to the model inference, which greatly improves the interpretability of the proposed model. The performance of the proposed model is demonstrated by experiments on a real-world online social network dataset. Experimental results show the proposed model achieves acceptable accuracy with both fully and partially available data. Comparison experiments were conducted, and the results show the proposed model’s promise for trust inference in some circumstances.

This paper focuses on consensus reaching process (CRP) under social network in which the trust relationship expressed by linguistic information. A new feedback mechanism in social network group decision making (SN-GDM) is proposed, which mainly consists of the following two aspects: (1) The propagation of distributed linguistic trust is investigated to study trust relation among experts; (2) A maximum self-esteem degree based feedback mechanism is developed to produce personalized advice for reaching higher group consensus. To do so, a novel linguistic trust propagation method is proposed to obtain the complete trust relationship among group. The self-esteem degree is used to define the extent that an individual makes concessions. Then, a maximum self-esteem degree based optimal feedback mechanism is built to produce personalized advice to help inconsistent experts make change of their opinion. Its novelty lies in the establishment of an optimization model with the nonlinear group self-esteem degree function as the objective function while group consensus threshold as the restrictions. Therefore, the inconsistent experts will reach a group consensus with the minimum loss of self-esteem degree, and then, it achieves the optimal balance between individual self-esteem and group consensus. Finally, a ranking process is applied to derive the appropriate consensus solution. • A new distributed linguistic trust propagation operator is built. • The concept of self-esteem degree (SED) is defined. • A maximum self-esteem degree based feedback mechanism is investigated. [ABSTRACT FROM AUTHOR]

New energy vehicle (NEV), an Eco-friendly innovation to alleviate the problems of energy scarcity and environmental pollution, is increasingly popular in many countries. Various new energy vehicles are provided with quantity of basic information (e.g., performance, quality, and price), which hinders potential users from effectively finding the most desired or interested new energy vehicles to satisfy their personalized requirements. This paper proposes a three-stage recommendation method for facilitating users to find the proper NEV considering users' preferences and social trust relationship. In the first stage, the users' preferences on evaluation criteria are determined by best-worst method (BWM) through hesitant fuzzy preference comparison vectors. In the second stage, the users' demographic similarity is obtained considering different formats of information, and then users' trust degrees are generated from the entire propagation paths using n dimensional path-ordering-induced order-weighted averaging (NP-IOWA) operator, thereby obtaining the trust-based similarity. In the third stage, the comprehensive user-rating matrix is constructed with the obtained weights, and then, it is combined with the trust-based similarity to recommend NEV based on collaborative filtering technique. A case study is given to illustrate the feasibility of the proposed method and the comparative analysis is conducted to show the advantages of the proposed method. [ABSTRACT FROM AUTHOR]

Cooperative offloading in mobile edge computing enables resource-constrained edge clouds to help each other with computation-intensive tasks. However, the power of such offloading could not be fully unleashed, unless trust risks in collaboration are properly managed. As tasks are outsourced and processed at the network edge, completion latency usually presents high variability that can harm the offered service levels. By jointly considering these two challenges, we propose OLCD, an Online Learning-aided Cooperative offloaDing mechanism under the scenario where computation offloading is organized based on accumulated social trust. Under co-provisioning of computation, transmission, and trust services, trust propagation is performed along the multi-hop offloading path such that tasks are allowed to be fulfilled by powerful edge clouds. We harness Lyapunov optimization to exploit the spatial-temporal optimality of long-term system cost minimization problem. By gap-preserving transformation, we decouple the series of bidirectional offloading problems so that it suffices to solve a separate decision problem for each edge cloud. The optimal offloading control can not materialize without complete latency knowledge. To adapt to latency variability, we resort to the delayed online learning technique to facilitate completion latency prediction under long-duration processing, which is fed as input to queued-based offloading control policy. Such predictive control is specially designed to minimize the loss due to prediction errors over time. We theoretically prove that OLCD guarantees close-to-optimal system performance even with inaccurate prediction, but its robustness is achieved at the expense of decreased stability. Trace-driven simulations demonstrate the efficiency of OLCD as well as its superiorities over prior related work. [ABSTRACT FROM AUTHOR]

Due to the urgent nature of emergency decision making, it is necessary to reach the consensus requirement quickly. Ordinal consensus measure explores the relation between the rankings and helps to intuitively know which alternative needs to be adjusted to accelerate the improvement of consensus. Moreover, decision makers (DMs) in the decision making problem are often connected through trust relationships which affect the DMs' judgments in the process of DMs' interaction. Therefore, this paper explores trust network-based group decision-making in which the consensus level is estimated by an ordinal consensus measure. We first focus on the supplementation of an incomplete trust network. One of the most common methods is to design the trust propagation operator, whereas the intensity of information propagation may be different in various scenarios. Therefore, considering the different numerical scale of the linguistic term set, a trust propagation operator with different intensity of trust propagation is designed to obtain the indirect trust relationship. In the process of supplementing the incomplete trust network, the contribution of DMs to propagating information is concerned, which can be described by the betweenness centrality, and the importance weights of DMs are determined by combining the betweenness centrality and trust in-degree. In the consensus reaching process, we first propose an improved ordinal consensus measure, which takes into account the consistency of orders of the same alternative in different rankings as well as the importance of positions of alternatives. Then, we design the identification rule and the feedback mechanism for those with low consensus levels. The identification rule is used to select the DMs which first few alternatives in the ranking are different with those in the ranking of group. And in the feedback mechanism, the referenced preference relation (FPR) obtained by the trust network is provided for the identified DMs. Afterwards, combining the referenced FPR, an optimization model is designed to give the adjustment opinion. Finally, a numerical example elaborates on the feasibility of the trust propagation operator and consensus model. The comparative analysis demonstrates the rationality and effectiveness of the proposed model. • A trust propagation operator with the intensity of trust propagation. • Considering the trust in-degree and betweenness centrality in DM's weights. • An ordinal consensus considered consistency of order and importance of position. • Considering the referenced information in the feedback mechanism. [ABSTRACT FROM AUTHOR]

According to the recommendation information processing problem in complex network environment,a trust model based on the recommendation chain classification was proposed.The classification method was based on honesty attribute of nodes,which could choose an effective recommendation chain on the basis of practical experience data.The recommendation information dissemination parameters were based on the information gain,which made recommendation information be more accurate.The factor of time was also considered in this model.The ability of interaction and the one of honesty were distinguished clearly.The concept of information entropy in information theory was used in the final ag-gregation calculation of direct trust and recommendation trust,which could get rid of the ambiguity of the previous sub-jective parameter settings.The main polymerization parameters could be continuously corrected with the interactions in order to achieve the situation being closest to the reality.Simulation results show the validity of recommendation chain classification and the rationality of the parameter settings in the proposed model.

Collaborative filtering (CF) is the most popular recommendation approach that has been extensively employed in recommender systems. However, it suffers from some weaknesses, including problems with cold start users, data sparsity and difficulty in detecting malicious users. Trust-based recommender systems can overcome these weaknesses by using the ratings of trusted users. However, since users often provide few trust statements, trust networks are typically sparse and therefore the cold start and sparsity problems still remain. In this paper, we use the positive correlation between trust and interest similarity to enrich trust network by similarity relations and propose a stochastic trust propagation-based method, called LTRS, which utilizes the enriched trust network to provide enhanced recommendations. In comparison with existing recommender systems combining trust and similarity information, the proposed system (1) incorporates both trust and similarity relations in the trust propagation process and, in this way, increases the coverage and accuracy of predictions; and (2) addresses the dynamic nature of both trust and similarity by modelling the enriched network as a stochastic graph, and continuously captures their variations during the recommendation process and not at fixed intervals. The experimental results indicate that the proposed method can significantly improve the recommendation accuracy and coverage. [ABSTRACT FROM AUTHOR]

Abstract The popularity of social networks has significantly promoted online individual interaction in the society. In online individual interaction, trust plays a critical role. It is very important to infer the trust among individuals, especially for those who have not had direct contact previously in social networks. In this paper, a restricted traversal method is defined to identify the strong trust paths from the truster and the trustee. Then, these paths are aggregated to predict the trust rate between them. During the traversal on a social network, interest topics and topology features are comprehensively considered, where weighted interest topics are used to measure the semantic similarity between users. In addition, trust propagation ability of users is calculated to indicate micro topology information of the social network. In order to find the t o p - k most trusted neighbors, two combination strategies for the above two factors are proposed in this paper. During trust inference, the traversal depth is constrained according to the heuristic rule based on the "small world" theory. Three versions of the trust rate inference algorithm are presented. The first algorithm merges interest topics and topology features into a hybrid measure for trusted neighbor selection. The other two algorithms consider these two factors in two different orders. For the purpose of performance analysis, experiments are conducted on a public and widely-used data set. The results show that our algorithms outperform the state-of-the-art algorithms in effectiveness. In the meantime, the efficiency of our algorithms is better than or comparable to those algorithms. [ABSTRACT FROM AUTHOR]