Data Mining - Concepts and Techniques.

Statistics for Spatial Data.

Even though human movement and mobility patterns have a high degree of freedom and variation, they also exhibit structural patterns due to geographic and social constraints. Using cell phone location data, as well as data from two online location-based social networks, we aim to understand what basic laws govern human motion and dynamics. We find that humans experience a combination of periodic movement that is geographically limited and seemingly random jumps correlated with their social networks. Short-ranged travel is periodic both spatially and temporally and not effected by the social network structure, while long-distance travel is more influenced by social network ties. We show that social relationships can explain about 10% to 30% of all human movement, while periodic behavior explains 50% to 70%. Based on our findings, we develop a model of human mobility that combines periodic short range movements with travel due to the social network structure. We show that our model reliably predicts the locations and dynamics of future human movement and gives an order of magnitude better performance than present models of human mobility.

/pdf/friendship-and-mobility-user-movement-in-location-based-2u0rrxuc9l.pdf

Friendship and mobility: user movement in location-based social networks

The theory of affordances

A great variety of systems in nature, society and technology -- from the web of sexual contacts to the Internet, from the nervous system to power grids -- can be modeled as graphs of vertices coupled by edges The network structure, describing how the graph is wired, helps us understand, predict and optimize the behavior of dynamical systems In many cases, however, the edges are not continuously active As an example, in networks of communication via email, text messages, or phone calls, edges represent sequences of instantaneous or practically instantaneous contacts In some cases, edges are active for non-negligible periods of time: eg, the proximity patterns of inpatients at hospitals can be represented by a graph where an edge between two individuals is on throughout the time they are at the same ward Like network topology, the temporal structure of edge activations can affect dynamics of systems interacting through the network, from disease contagion on the network of patients to information diffusion over an e-mail network In this review, we present the emergent field of temporal networks, and discuss methods for analyzing topological and temporal structure and models for elucidating their relation to the behavior of dynamical systems In the light of traditional network theory, one can see this framework as moving the information of when things happen from the dynamical system on the network, to the network itself Since fundamental properties, such as the transitivity of edges, do not necessarily hold in temporal networks, many of these methods need to be quite different from those for static networks

Temporal Networks

Sequence, widely appearing in various applications (e.g. event logs, text documents, etc) is an ordered list of objects. Exploring correlated objects in a sequence can provide useful knowledge among the objects, e.g., event causality in event log and word phrases in documents. In this paper, we introduce correlation querythat finds correlated pairs of objects often appearing closely to each other in a given sequence. A correlation query is specified by two control parameters, distance bound, the requirement of object closeness, and correlation threshold, the minimum requirement of correlation strength of result pairs. Instead of processing the query by scanning the sequence multiple times, that is called Multi-Scan Algorithm (MSA), we propose One-Scan Algorithm (OSA)and Index-Based Algorithm (IBA). OSA accesses a queried sequence once and IBA considers correlation threshold in the execution and effectively eliminates unneeded candidates from detail examination. An extensive set of experiments is conducted to evaluate all these algorithms. Among them, IBA, significantly outperforming the others, is the most efficient.

Searching Correlated Objects in a Long Sequence

Peer-to-peer (P2P) systems have received a lot of attention due to the popularity of applications such as SETI, Napster, Gnutella, and Morpheus. The P2P systems present tremendous challenges in search- ing data items among the numerous host nodes. While search has been studied in a similar but different con- text, i.e., parallel and distributed database systems, the large scale and dynamic membership change of P2P systems require the search issue to be re-examined. Existing search techniques in unstructured peer-to-peer overlay networks incur excessive network traffic. In this paper, we investigate the issues of trading-off stor- age space at peers to reduce network overhead in unstructured P2P overlay networks. We propose to use sig- natures for directing searches, and introduce three schemes, namely complete-neighborhood signature (CN), partial-neighborhood superimposed signature (PN-S), and partial-neighborhood appended signature (PN-A), to facilitate efficient searching of shared content in P2P networks. With little storage overhead, these signa- tures improve the performance of content search and thus significantly reduce the volume of network traffic. Extensive analysis and simulations are conducted to evaluate the performance of our proposal with existing P2P content search methods, including Gnutella, random walk, and local index. Results show that PN-A gives the best performance at a small storage cost.

http://www.cse.psu.edu/~wul2/Publications/wlee%20JOC_SE3_2.pdf

Performance Evaluation of Neighborhood Signature Techniques for Peer-to-Peer Search

Most social networks exhibit community structures, in which nodes are tightly connected to each other within a community but only loosely connected to nodes in other communities. Researches on community mining have received a lot of attention; however, most of them are based on a centralized system model and thus not applicable to the distributed model of P2P networks. In this paper, we propose a distributed community mining algorithm, namely Asynchronous Clustering and Merging scheme (ACM), for computing environments. Due to the dynamic and distributed nature of P2P networks, The ACM scheme employs an asynchronous strategy such that local clustering is executed without requiring an expensive global clustering to be performed in a synchronous fashion. Experimental results show that ACM is able to discover community structures with high quality while outperforming the existing approaches.

https://ir.nctu.edu.tw:443/bitstream/11536/1153/1/000263157700044.pdf

Mining Community Structures in Peer-to-Peer Environments

While the K-Nearest-Neighbor (KNN) problem is well studied in the traditional wired, disk-based client-server environment, it has not been tackled in a wireless broadcast environment. In this paper, the problem of organizing location dependent data and answering KNN queries on air are investigated. The linear property of wireless broadcast media and power conserving requirement of mobile devices make this problem particularly interesting and challenging. An efficient data organization, called sorted list, and the corresponding search algorithm are proposed and compared with the well-known spatial index, R-Tree. In addition, we develop an approximate search scope to guide the search at the very beginning of the search process and a learning algorithm to adapt the search scope during the search to improve energy and access efficiency. Simulation based performance evaluation is conducted to compare sorted list and R-Tree. The results show that the utilization of search scope and learning algorithm improves search efficiency of both index mechanisms significantly. While R-Tree is more power efficient when a large number of nearest neighbors is requested, the sorted list has better access efficiency and less power consumption when the number of nearest neighbors is small.

Search K nearest neighbors on air

Geospatial tagging (geotagging) is an emerging and very promising application that can help users find a wide variety of location-specific information, and thereby facilitate the development of advanced location-based services. Conventional geotagging systems share some limitations, such as the use of a two-phase operating model and the tendency to tag popular objects with simple contexts. To address these problems, a number of geotagging systems based on the concept of `Games with a Purpose' (GWAP) have been developed recently. In this study, we use analysis to investigate these new systems. Based on our analysis results, we design three metrics to evaluate the system performance, and develop five task assignment algorithms for GWAP-based systems. Using a comprehensive set of simulations under both synthetic and realistic mobility scenarios, we find that the Least-Throughput-First Assignment algorithm (LTFA) is the most effective approach because it can achieve competitive system utility, while its computational complexity remains moderate. We also find that, to improve the system utility, it is better to assign as many tasks as possible in each round. However, because players may feel annoyed if too many tasks are assigned at the same time, it is recommended that multiple tasks be assigned one by one in each round in order to achieve higher system utility.

Wang-Chien Lee

Papers

Searching Correlated Objects in a Long Sequence

Performance Evaluation of Neighborhood Signature Techniques for Peer-to-Peer Search

Mining Community Structures in Peer-to-Peer Environments

Search K nearest neighbors on air

The Design and Evaluation of Task Assignment Algorithms for GWAP-based Geospatial Tagging Systems