Yongkoo Han

A Framework for Supervising Lifestyle Diseases Using Long-Term Activity Monitoring

Activity monitoring of a person for a long-term would be helpful for controlling lifestyle associated diseases. Such diseases are often linked with the way a person lives. An unhealthy and irregular standard of living influences the risk of such diseases in the later part of ones life. The symptoms and the initial signs of these diseases are common to the people with irregular lifestyle. In this paper, we propose a novel healthcare framework to manage lifestyle diseases using long-term activity monitoring. The framework recognizes the users activities with the help of the sensed data in runtime and reports the irregular and unhealthy activity patterns to a doctor and a caregiver. The proposed framework is a hierarchical structure that consists of three modules: activity recognition, activity pattern generation and lifestyle disease prediction. We show that it is possible to assess the possibility of lifestyle diseases from the sensor data. We also show the viability of the proposed framework.

Confident wrapper-type semi-supervised feature selection using an ensemble classifier

Feature selection is an important data preprocessing step in pattern recognition. Recently, a wrapper-type semi-supervised feature selection method, known as FW-SemiFS, was proposed to overcome the small labeled sample problem of supervised feature selection. FW-SemiFS does not consider the confidence of predicted unlabeled data, but rather evaluates the relevance of features according to their frequency. Such frequencies are obtained via iterative supervised sequential forward feature selection (SFFS). However, the large amount of computational time associated with iterative SFFS is detrimental to FW-SemiFS. Furthermore, this relevance evaluation method eliminates the primary advantage of wrapper-type feature selection: the ability to evaluate the discriminative power of a combination of features. In this paper, we propose a new wrapper-type semi-supervised feature selection framework that can select a more relevant feature subset using confident unlabeled data. The proposed framework, called ensemble-based semi-supervised feature selection (EN-SemiFS), employs an ensemble classifier that supports the estimation of the confidence of unlabeled data. We analyzed the relationship between wrapper-type feature selection and the confidence of unlabeled data and explored how this relationship can make the semisupervised feature selection framework faster and more accurate. The experimental results revealed that the proposed method can select a more relevant feature subset when compared to existing methods.

An Efficient Method for Computing Similarity Between Frequent Subgraphs

Frequent sub graph mining and graph similarity measures are fundamental and prominent graph analytical techniques. These techniques are often applied together in many graph mining techniques such as clustering and classification. However, these techniques suffer from long running times because frequent sub graph mining and graph similarity measures have been applied independently. In this paper, we propose an efficient method that measures similarity between frequent sub graphs. Our method exploits byproducts of frequent sub graph mining for avoiding costly common sub graph search required in similarity measures. Through experiments on real world graph data, we show that our method measures similarities among all pair of frequent sub graphs within practical time.

Initial training data selection for active learning

The crucial issue in many classification applications is how to achieve the best possible classifier with a limited number of labeled training data. Active learning is one method which addresses this issue by selecting the most informative data for training. In this work, we argue that the performance of active learning could be improved through carefully selecting the initial training samples. To confirm our argument, we propose three initial training data selection mechanisms based on fuzzy clustering method: center-based selection, border-based selection and hybrid selection. Center-based selection selects the samples with high degree of membership in each cluster as initial training data. Border-based selection selects the samples around the border between clusters. Hybrid selection is the combination of center-based selection and border-based selection. The effects of them are empirically studied on a set of UCI data sets. Experimental result indicates that, compared with randomly selecting initial training samples, hybrid selection can effectively enhance the performance of active learning.

Distance-Constraint k-Nearest Neighbor Searching in Mobile Sensor Networks

The k-Nearest Neighbors (kNN) query is an important spatial query in mobile sensor networks. In this work we extend kNN to include a distance constraint, calling it a l-distant k-nearest-neighbors (l-kNN) query, which finds the k sensor nodes nearest to a query point that are also at l or greater distance from each other. The query results indicate the objects nearest to the area of interest that are scattered from each other by at least distance l. The l- kNN query can be used in most kNN applications for the case of well distributed query results. To process an l-kNN query, we must discover all sets of kNN sensor nodes and then find all pairs of sensor nodes in each set that are separated by at least a distance l. Given the limited battery and computing power of sensor nodes, this l-kNN query processing is problematically expensive in terms of energy consumption. In this paper, we propose a greedy approach for l- kNN query processing in mobile sensor networks. The key idea of the proposed approach is to divide the search space into subspaces whose all sides are l. By selecting k sensor nodes from the other subspaces near the query point, we guarantee accurate query results for l- kNN. In our experiments, we show that the proposed method exhibits superior performance compared with a post-processing based method using the kNN query in terms of energy efficiency, query latency, and accuracy.

RDB2RDF: completed transformation from relational database into RDF ontology

One of the most advantages of the Semantic Web is to augment the data with a well-defined meaning and linking between data by using the RDF ontology language. Today most of data are stored in relational databases. In order to reuse and infer this data on the Semantic Web, there is a need for converting the data stored in relational databases to the form of RDF. Some approaches have been proposed, however, most of them transform a single table into RDF triples. This paper presents RDB2RDF, a complete method to transform all tables in the relational database into RDF ontology. The transformation makes it possible to reverse RDF ontology to relational tables. Most of all, all the steps in RDB2RDF are done automatically without any user intervention.

Disk-based shortest path discovery using distance index over large dynamic graphs

Abstract The persistent alternation of the internet world is changing networks rapidly. Shortest path discovery, especially over dynamic networks such as web page links, telephone or route networks, and ontologies, has received intense attention because of its importance for services in IoT. For example, when a new road is newly opened or becomes unavailable for any unexpected reason, the shortest paths must be recomputed. The system should respond promptly to its users with the updated recommended paths. In this paper, we propose a disk-based shortest path method that updates the shortest paths in a very large dynamic graph efficiently. The proposed method uses partial shortest paths as indices for efficient shortest path discovery. We classify the changes in the graph into four cases, such as the insertion or deletion of edges and the increase or decrease of edge weights. Our proposed strategy considers updating only the corresponding parts of the indices for each case. Our experiments on real-world dynamic datasets verify that the proposed framework updates the shortest paths 4 to 50 times faster than the existing type of framework.

iTri: Index-based triangle listing in massive graphs

Abstract Triangle listing is a basic operator when dealing with many graph problems. However, in-memory algorithms do not work well with recently developed massive graphs such as social networks because these graphs cannot be accommodated in the memory. Thus, external memory-based algorithms have been proposed recently, but these approaches still require frequent multiple scans of the whole graph on the disk and large volumes of calculations are performed that involve the whole graph during every iteration. In this study, we propose a novel index-based method for listing triangles in massive graphs. First, we present new notions for the vertex range index and potential cone vertex index. Next, we propose an index join-based triangle listing algorithm. Our method accesses the indexed data asynchronously and joins them to list triangles using a multi-threaded parallel processing technique. Based on experiments, we demonstrate that our algorithm outperforms the state-of-the-art solution methods by three to eight times in terms of the wall clock time.

Efficient routing on finding recommenders for trust-aware recommender systems

The trust-aware recommender system (TARS) is a newly proposed trust-aware application. It is able to solve the data sparseness problem of the conventional recommender systems. One of the basic research challenges in TARS is to find the recommenders efficiently. Existing works of TARS use the strategy of random walk to find the recommenders, which is obviously low efficiency. Though the trust network has been verified to be the scale-free network, due to the small power of its degree distributions, we have verified via experiments that the prediction coverage of TARS is very limited by applying the classical routing protocol of scale-free networks directly. We therefore propose a routing protocol for TARS, which is able to efficiently find reliable recommenders for the users of TARS. Our protocol is able to achieve much higher prediction coverage than the classical routing protocol of scale-free networks, while the computational complexity is greatly reduced comparing with existing works of TARS.

More reputable recommenders give more accurate recommendations

Existing models of the Trust-Aware Recommender System (TARS) build personalized trust networks for the active users to predict ratings. These models have reasonable rating prediction performances, while suffer from high computational complexity. One solution is to utilize the global rating prediction mechanism for TARS, in which an intuitive assumption is that more reputable recommenders give more accurate recommendations. In addition, due to the scale-freeness of the trust network, some users have and continuously have superior reputations than others. However, we show via comprehensive experiments on the real TARS data that the recommendations given by recommenders with higher reputations do not tend to be more accurate. Furthermore, even the recommendations given by the recommenders with superior high reputations do not tend to more accurate. Our experimental study provides promising directions for the future research on the rating prediction mechanism of TARS.