Wenjia Niu

A DeepWalk-Based Approach to Defend Profile Injection Attack in Recommendation System

In the open social networks, the analysis of user data after the injection attack has a great impact on the recommendation system. K-Nearest Neighbor-based collaborative filtering algorithms are very vulnerable to this attack. Another recommendation algorithm based on probabilistic latent semantic analysis has relatively accurate recommendation, but it is not very stable and robust against attacks on the overall user data of the recommendation system. Here is used to DeepWalk the user network processing, while taking advantage of the user profile feature time series to consider the user’s behavior over time, the algorithm also analyzes the stability and robustness of DeepWalk and user profile. The results show that especially the DeepWalk-based approach can achieve comparable recommendation accuracy.

Beyond the Aggregation of Its Members—A Novel Group Recommender System from the Perspective of Preference Distribution

This paper focuses on recommending items to group of users rather than individual users. To model group profile, existing researches almost aggregate preferences of members into a single value, and thus cannot reflect actual group profile of groups with conflicting characteristics. Therefore, we propose a novel group recommender system mechanism. It views group profile as preference distribution, and then models item recommendation process as a multi-criteria decision making process, in order to obtain better recommendation results. Finally, experiments are conducted to verify the proposed approach.

A Hidden Astroturfing Detection Approach Base on Emotion Analysis

This paper aims to take detection of hidden astroturfing based on emotion analysis. We propose a hidden astroturfing detection method which combines emotion analysis and unfair rating detection together. This approach contains five functional modules as: a data crawling module, pre-processing module, bag-of-word establishment module, emotion mining and analysis module and matching module. We give ROC curve (AUC) to evaluate the approach proposed in this paper. The results show that this method can realize the detection of implicit astroturfing under the prerequisite of improving the emotion classification accuracy. Our work discovers and studies a new hidden astroturfing characteristic, and construct a corpus manually for text emotion classification that establish a basis for our future research.

A Method to Effectively Detect Vulnerabilities on Path Planning of VIN

Reinforcement Learning has been used on path planning for a long time, which is thought to be very effective, especially the Value Iteration Networks (VIN) with strong generalization ability. In this paper, we analyze the path planning of VIN and propose a method that can effectively find vulnerable points in VIN. We build a 2D navigation task to test our method. The experiment for interfering VIN is conducted for the first time. The experimental results show that our method has good performance on finding vulnerabilities and could automatically adding obstacles to obstruct VIN path planning.

A Cross-Modal CCA-Based Astroturfing Detection Approach

In recent years, astroturfing can generate abnormal, damaging even illegal behaviors in cyberspace which may mislead the public perception and bring a bad effect on both Internet users and society. This paper aims to design a algorithm to detect astroturfing in online shopping effectively and help users to identify potential online astroturfers quickly. The previous work used single method text-text or image-image to detect astroturfing, while in this paper we first propose a cross-modal canonical correlation analysis model (CCCA) which combines text and images. First, we identify several features of astroturfing and analysis these features. Then, we use feature extraction algorithm, image similarity algorithm and CCA algorithm, and propose a cross-modal method to detect astroturfing which release comments with pictures. We also conduct an experiment on a Taobao dataset to verify our method. The experimental results show that the supervised method proposed is effective.