Yingyuan Xiao

An algorithm on fairness verification of mobile sink routing in wireless sensor network

Congestion and starvation will occur among some nodes due to the emerging serious unfairness, which is derived from the limited communication capabilities of all nodes and sink or in the case of a mobile sink moving to a new place. The problem to be solved is to balance the network and keep the fairness for all nodes. For this purpose, this paper focuses on verifying the fairness of mobile sink routing based on both state and action, which is realized mainly by composing Labeled Kripke Transition Systems (LKTS). First, an approach is presented by LKTS to model node behaviors. Second, a notion of Fair Computational Tree Logic (CTL) is introduced to describe the fairness formulae in branching time transitions, and four kinds of fairness assumptions are defined for fairness verification. Moreover, in order to avoid the problem of state-space explosion, Bounded model Checking to explore states and transitions on-the-fly until a witness is found, while Strong Connected Components algorithm is used to pick up fair paths under fairness constraints of Fair CTL. The experimental results show the superiority of our method by the savings in memory and time consumptions during the mobile sink routing process.

Leveraging communication information among readers for RFID data cleaning

Radio Frequency Identification (RFID) technologies are used in many applications for data collection. However, raw RFID readings are usually of low quality due to frequent occurrences of false negative, false positive and duplicate readings. A number of RFID data cleaning techniques are proposed to solve the problem. In this paper we explore to use communication information for RFID data cleaning and make RFID readers produce less dirty data at the early stage. First, we devise a reader communication protocol for efficiently utilizing the communication information among readers. Then, the cell event sequence tree with parameters is proposed. Finally, we present three novel RFID data cleaning methods, respectively for duplicate readings, false positive readings and data interpolating. To the best of our knowledge, this is the first work utilizing the communication information among readers in RFID data cleaning. We conduct extensive experiments, and the experimental results demonstrate the feasibility and effectiveness of our methods.

Data Interpolating over RFID Data Streams for Missed Readings

While tracing objects or analyzing human activities with RFID data sets, the quality of RFID data is a crucial aspect. The raw RFID data streams, however, tend to be noisy, including missed readings and unreliable readings. Traditional data cleaning tends to focus on a small set of well-defined tasks, including transformation, matching, and duplicate elimination. In this paper, we focus on exploring efficient methods for interpolating missed readings. We propose a novel probabilistic interpolating method and three novel deterministic interpolating methods based on time interval, containment relationship and inertia of objects, respectively. We conduct extensive experiments and the experimental results demonstrate the feasibility and effectiveness of our methods.

A Personalized News Recommendation System Based on Tag Dependency Graph

The tags of news articles give readers the most important and relevant information regarding the news articles, which are more useful than a simple bag of keywords extracted from news articles. Moreover, latent dependency among tags can be used to assign tags with different weight. Traditional content-based recommendation engines have largely ignored the latent dependency among tags. To solve this problem, we implemented a prototype system called PRST, which is presented in this paper. PRST builds a tag dependency graph to capture the latent dependency among tags. The demonstration shows that PRST makes news recommendation more effectively.

Incorporating Contextual Information into a Mobile Advertisement Recommender System

The ever growing popularity of smart mobile devices and rapid advent of wireless technology have given rise to a new class of advertising system, i.e., mobile advertisement recommender system. The traditional internet advertising systems have largely ignored the fact that users interact with the system within a particular “context”. In this paper, we implemented a mobile advertisement recommender prototype system called MARS. MARS captures different user’s contextual information to improve recommendation results. The demonstration shows that MARS makes advertisement recommendation more effectively.

Searching Desktop Files Based on Access Logs

People often meet trouble in searching a desktop file when they can not remember exact words of its filename. In this paper, we firstly propose an algorithm to generate access logs by monitoring desktop operations and implement a prototype. By running it in several computers of selected participants we collected a data set of access logs. Then we propose a graph model to represent personal desktop files and their relationships, and highlight two file relationships(content relationship and time relationship) to help users search desktop files. Based on the graph model, we propose a desktop search method, and the experimental results show the feasibility and effectiveness of our methods.

Ontology Driven Formal Description of Swift Trust and Its Computational Model

This paper builds up an ontology conceptual model for swift trust, which is reasonable when considering the fuzzy nature of trust, as well as the heterogeneity of diverse ontologys developed by various engineers with different knowledge structure. Next, D-S theory is chosen as the reasoning tool for our swift trust, which may be more appropriate than the other uncertain theories including Fuzzy sets theory and Bayesian method. Finally, examples are taken to verify the validity and correctness of our proposed model.