Dik Lun Lee

Exploiting geographical influence for collaborative point-of-interest recommendation

In this paper, we aim to provide a point-of-interests (POI) recommendation service for the rapid growing location-based social networks (LBSNs), e.g., Foursquare, Whrrl, etc. Our idea is to explore user preference, social influence and geographical influence for POI recommendations. In addition to deriving user preference based on user-based collaborative filtering and exploring social influence from friends, we put a special emphasis on geographical influence due to the spatial clustering phenomenon exhibited in user check-in activities of LBSNs. We argue that the geographical influence among POIs plays an important role in user check-in behaviors and model it by power law distribution. Accordingly, we develop a collaborative recommendation algorithm based on geographical influence based on naive Bayesian. Furthermore, we propose a unified POI recommendation framework, which fuses user preference to a POI with social influence and geographical influence. Finally, we conduct a comprehensive performance evaluation over two large-scale datasets collected from Foursquare and Whrrl. Experimental results with these real datasets show that the unified collaborative recommendation approach significantly outperforms a wide spectrum of alternative recommendation approaches.

Document ranking and the vector-space model

Efficient and effective text retrieval techniques are critical in managing the increasing amount of textual information available in electronic form. Yet text retrieval is a daunting task because it is difficult to extract the semantics of natural language texts. Many problems must be resolved before natural language processing techniques can be effectively applied to a large collection of texts. Most existing text retrieval techniques rely on indexing keywords. Unfortunately, keywords or index terms alone cannot adequately capture the document contents, resulting in poor retrieval performance. Yet keyword indexing is widely used in commercial systems because it is still the most viable way by far to process large amounts of text. Using several simplifications of the vector-space model for text retrieval queries, the authors seek the optimal balance between processing efficiency and retrieval effectiveness as expressed in relevant document rankings.

Location-based spatial queries

In this paper we propose an approach that enables mobile clients to determine the validity of previous queries based on their current locations. In order to make this possible, the server returns in addition to the query result, a validity region around the clients location within which the result remains the same. We focus on two of the most common spatial query types, namely nearest neighbor and window queries, define the validity region in each case and propose the corresponding query processing algorithms. In addition, we provide analytical models for estimating the expected size of the validity region. Our techniques can significantly reduce the number of queries issued to the server, while introducing minimal computational and network overhead compared to traditional spatial queries.

A generic framework for monitoring continuous spatial queries over moving objects

This paper proposes a generic framework for monitoring continuous spatial queries over moving objects. The framework distinguishes itself from existing work by being the first to address the location update issue and to provide a common interface for monitoring mixed types of queries. Based on the notion of safe region, the client location update strategy is developed based on the queries being monitored. Thus, it significantly reduces the wireless communication and query reevaluation costs required to maintain the up-to-date query results. We propose algorithms for query evaluation/reevaluation and for safe region computation in this framework. Enhancements are also proposed to take advantage of two practical mobility assumptions: maximum speed and steady movement. The experimental results show that our framework substantially outperforms the traditional periodic monitoring scheme in terms of monitoring accuracy and CPU time while achieving a close-to-optimal wireless communication cost. The framework also can scale up to a large monitoring system and is robust under various object mobility patterns.

Data management in location-dependent information services

Location-dependent information services have great promise for mobile and pervasive computing environments. They can provide local and nonlocal news, weather, and traffic reports as well as directory services. Before they can be implemented on a large scale, however, several research issues must be addressed.

Cache invalidation and replacement strategies for location-dependent data in mobile environments

Mobile location-dependent information services (LDISs) have become increasingly popular in recent years. However, data caching strategies for LDISs have thus far received little attention. In this paper, we study the issues of cache invalidation and cache replacement for location-dependent data under a geometric location model. We introduce a new performance criterion, called caching efficiency, and propose a generic method for location-dependent cache invalidation strategies. In addition, two cache replacement policies, PA and PAID, are proposed. Unlike the conventional replacement policies, PA and PAID take into consideration the valid scope area of a data value. We conduct a series of simulation experiments to study the performance of the proposed caching schemes. The experimental results show that the proposed location-dependent invalidation scheme is very effective and the PA and PAID policies significantly outperform the conventional replacement policies.

IR-Tree: An Efficient Index for Geographic Document Search

Given a geographic query that is composed of query keywords and a location, a geographic search engine retrieves documents that are the most textually and spatially relevant to the query keywords and the location, respectively, and ranks the retrieved documents according to their joint textual and spatial relevances to the query. The lack of an efficient index that can simultaneously handle both the textual and spatial aspects of the documents makes existing geographic search engines inefficient in answering geographic queries. In this paper, we propose an efficient index, called IR-tree, that together with a top-k document search algorithm facilitates four major tasks in document searches, namely, 1) spatial filtering, 2) textual filtering, 3) relevance computation, and 4) document ranking in a fully integrated manner. In addition, IR-tree allows searches to adopt different weights on textual and spatial relevance of documents at the runtime and thus caters for a wide variety of applications. A set of comprehensive experiments over a wide range of scenarios has been conducted and the experiment results demonstrate that IR-tree outperforms the state-of-the-art approaches for geographic document searches.

Server Ranking for Distributed Text Retrieval Systems on the Internet

Keyword-based search services have become necessary tools for nding information resources on the Internet today. The rapid growth of information on the Internet renders centralized keyword index services incapable of collecting comprehensive resource meta-data in a timely manner. We argue that delegating the task of meta-data collection to local index servers is a more scalable approach. We propose a mechanism for integrating distributed autonomous index servers into a cooperative resource discovery system. Focusing on the retrieval eeec-tiveness of the system, we propose a set of methods , called CVV-based methods, for ranking and selecting index servers with respect to a query, and merging the results returned by the index servers. Through experiments, we evaluate the eeectiveness of the CVV-based methods, and compare our server ranking method with methods proposed by other researchers .

Using Signature Techniques for Information Filtering in Wireless and Mobile Environments

This paper discusses the issue of power conservation on mobile clients, e.g., palmtop, in wireless and mobile environments. It suggests that techniques using signatures are suitable for realtime information filtering on mobile clients. Three signature-based approaches, namely simple signature, integrated signature and multi-level signature schemes, are presented. The cost models for the access time and tune-in time of these three approaches are developed. We show that the multi-level signature method is in general better than the other two methods.

Semantic Caching in Location-Dependent Query Processing

A method is presented in this paper for answering location-dependent queries in a mobile computing environment. We investigate a common scenario where data objects (e.g., restaurants and gas stations) are stationary while clients that issue queries about the data objects are mobile. Our proposed technique constructs a Voronoi Diagram (VD) on the data objects to serve as an index for them. A VD defines, for each data object d, the region within which d is the nearest point to any mobile client within that region. As such, the VD can be used to answer nearest-neighbor queries directly. Furthermore, the area within which the answer is valid can be computed. Based on the VD, we develop a semantic caching scheme that records a cached item as well as its valid range. A simulation is conducted to study the performance of the proposed semantic cache in comparison with the traditional cache and the baseline case where no cache is used. We show that the semantic cache has a much better performance than the other two methods.