Factorization machines and deep views-based co-training for improving answer quality prediction in online health expert question-answering services

Intelligent crowd sensing pickpocketing group identification using remote sensing data for secure smart cities

As a public infrastructure service, remote sensing data provided by smart cities will go deep into the safety field and realize the comprehensive improvement of urban management and services. However, it is challenging to detect criminal individuals with abnormal features from massive sensing data and identify groups composed of criminal individuals with similar behavioral characteristics. To address this issue, we study two research aspects: pickpocketing individual detection and pickpocketing group identification. First, we propose an IForest-FD pickpocketing individual detection algorithm. The IForest algorithm filters the abnormal individuals of each feature extracted from ticketing and geographic information data. Through the filtered results, the factorization machines (FM) and deep neural network (DNN) (FD) algorithm learns the combination relationship between low-order and high-order features to improve the accuracy of identifying pickpockets composed of factorization machines and deep neural networks. Second, we propose a community relationship strength (CRS)-Louvain pickpocketing group identification algorithm. Based on crowdsensing, we measure the similarity of temporal, spatial, social and identity features among pickpocketing individuals. We then use the weighted combination similarity as an edge weight to construct the pickpocketing association graph. Furthermore, the CRS-Louvain algorithm improves the modularity of the Louvain algorithm to overcome the limitation that small-scale communities cannot be identified. The experimental results indicate that the IForest-FD algorithm has better detection results in Precision, Recall and F1score than similar algorithms. In addition, the normalized mutual information results of the group division effect obtained by the CRS-Louvain pickpocketing group identification algorithm are better than those of other representative methods.

A Two-Phase Deep Learning-Based Recommender System: Enhanced by a Data Quality Inspector

Research regarding collaborative filtering recommenders has grown fast lately. However, little attention has been paid to discuss how the input data quality impacts the result. Indeed, some review-rating pairs that a user gave to an item are inconsistent and express a different opinion, making the recommendation result biased. To solve the above drawback, this study proposes a two-phase deep learning-based recommender system. Firstly, a sentiment predictor of textual reviews is created, serving as the quality inspector that cleans and improves the input for a recommender. To build accurate predictors, this phase tries and compares a set of deep learning-based algorithms. Secondly, besides only exploiting the consistent review-rating pairs generated by the quality inspector, this phase builds deep learning-based recommender engines. The experiments on a real-world dataset showed the proposed data quality inspector, based on textual reviews, improves the overall performance of recommenders. On average, applying deep learning-based quality inspectors result in an above 6% improvement in RMSE, and more than a 2% boost in F1 score, and accuracy. This is robust evidence to prove the importance of the input data cleaning process in this field. Moreover, empirical evidence indicates the deep learning approach is suitable for modeling the sentiment predictor, and the core recommendation process, clearly outperforming the traditional machine learning methods.