Research on predicting movements of mobile users has attracted a lot attentions in recent years. Many those prediction techniques are developed based only geographic features users' trajectories. In this paper, we propose novel approach for the next location user's movement both and semantic The core idea our model is cluster-based strategy which evaluates user frequent behaviors similar same cluster determined by analyzing common behavior Through comprehensive evaluation experiments,...

In recent years, research on measuring trajectory similarity has attracted a lot of attentions. Most similarities are defined based the geographic features mobile users' trajectories. However, trajectories geographically close may not necessarily be similar because activities implied by nearby landmarks they pass through different. this paper, we argue that better measurement should have taken into account semantics propose novel approach for recommending potential friends semantic...

In recent years, research on location predictions by mining trajectories of users has attracted a lot attention. Existing studies this topic mostly treat such as just type recommendation, that is, they predict the next user using recommenders. However, an usually visits somewhere for reasons other than interestingness. article, we propose novel mining-based prediction approach called Geographic-Temporal-Semantic-based Location Prediction (GTS-LP), which takes into account user's...

In recent years, researches on recommendation of urban Points-Of-Interest (POI), such as restaurants, based social information have attracted a lot attention. Although number social-based techniques been proposed in the literature, most their concepts are only individual or friends' check-in behaviors. It leads to that recommended POIs list is usually constrained within users' living area. Furthermore, since context-aware and environmental changes quickly, especially areas, how extract...

In recent years, research into the mining of user check-in behavior for point-of-interest (POI) recommendations has attracted a lot attention. Existing studies on this topic mainly treat such in traditional manner—that is, they POIs as items and check-ins ratings. However, users usually visit place reasons other than to simply say that have visited. article, we propose an approach referred Urban POI-Walk (UPOI-Walk), which takes account user's social-triggered intentions (SI),...

Computational intelligence has been used in many applications the fields of health sciences and epidemiology. In particular, owing to sudden massive spread COVID-19, researchers around globe have devoted intensive efforts into development computational methods systems for combating pandemic. Although there more than 200,000 scholarly articles on SARS-CoV-2, other related coronaviruses, these did not specifically address in-depth key issues applying combat COVID-19. Hence, it would be...

In recent years, fraud is increasing rapidly with the development of modern technology and global communication. Although many literatures have addressed detection problem, these existing works focus only on formulating problem as a binary classification problem. Due to limitation information provided by telecommunication records, such classifier-based approaches for fraudulent phone call normally do not work well. this paper, we develop graph-mining-based framework mobile application...

The aim of this study was to assess the association between air pollutant exposure and interstitial lung disease (ILD) in patients with connective tissue diseases (CTDs).A nationwide, population-based, matched case-control Taiwan.Using 1997-2013 Taiwanese National Health Insurance Research Database, we identified newly diagnosed CTD during 2001-2013, including systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), sclerosis (SSc), dermatomyositis (DMtis)/polymyositis (PM) primary...

Classification is an important and well-known technique in the field of machine learning, training data will significantly influence classification accuracy. However, real-world applications often are imbalanced class distribution. It to select suitable for distribution problem. In this paper, we propose a cluster-based sampling approach selecting representative as improve accuracy investigate effect under-sampling methods experiments, evaluate performances our other previous studies.

Social insurance plays crucial role for protecting the social functionality. Since different occupation would lead to level salary and risk, people is usually designed by characteristic of their occupation. In Taiwan, Ministry Health Welfare provides health care citizens guarantees basic income elderly life. Accordingly, operates sponsor many types insurances such as National Pension Insurance, Labor etc. However, due Population ageing, are going encounter crisis pension bankruptcy....

In recent years, location-based social networks (LBSNs) have received high attention. While this new breed of is nascent, there no large-scale analysis conducted to investigate the associations among users in locales network. paper, we propose four locale based metrics, including Locale Clustering Coefficient, Inward Transitivity, Assortativity and Assortability Coefficient make association on EveryTrail, a popular LBSN specialized sharing trips. Based result, observe that people who share...

In recent years, the Taiwan government has been calling for use of public transportation and popularizing pollution-reducing green vehicles. Passenger transport operators are being encouraged to replace traditional buses with electric buses, increase their in urban transportation. Reduced energy consumption operating costs important operational benefits passenger operators, driving behavior a significant impact on fuel consumption. Although many literatures or real-world systems have...

To assess the association of severe pulmonary arterial hypertension (PAH) with particulate matter <2.5 μm (p.m.2.5) and clinical data in patients systemic autoimmune rheumatic diseases (SARDs).We used 2003-2017 nationwide Taiwan to identify SARDs, including lupus erythematosus, rheumatoid arthritis, sclerosis, dermatomyositis/polymyositis primary Sjögren's syndrome. We identified 479 cases PAH selected controls matched (1:4) for age, sex, index year. conditional logistic regression analysis...

In recent years, telecommunication fraud has become more rampant internationally with the development of modern technology and global communication. Because rapid growth in volume call logs, task fraudulent phone detection is confronted big data issues real-world implementations. Although our previous work, FrauDetector , addressed this problem achieved some promising results, it can be further enhanced because focuses only on accuracy, whereas efficiency scalability are not top priorities....

Sensory navigation device is an important trend in the field of machine learning and data science. Nowadays, more sensory devices are built for blind people. The core such people usually implemented by Image Recognition Method. To build image recognition model, many tools online platforms proposed. However, these or not able to completely satisfy requirements device. a with satisfying people, ability reducing cost model training capability user-centric two main issues. Therefore, address...

Route planning satisfied multiple requests is an emerging branch in the route field and has attracted significant attention from research community recent years. The prevailing studies focus only on seeking a by minimizing single kind of Travel Cost, such as trip time or distance, among others. In reality, most users would like to choose appropriate route, neither fastest nor shortest route. Usually, user may have requirements, satisfy all requirements requested user. fact, could be...

Researches on recommending followees in social networks have attracted a lot of attentions recent years. Existing studies this topic mostly treat kind recommendation as just type friend recommendation. However, apart from making friends, the reason user to follow someone is inherently satisfy his/her information needs asymmetrical manner. In paper, we propose novel mining-based approach named Geographic-Textual-Social Based Followee Recommendation (GTS-FR), which takes into account...

In recent years, researches on recommendation systems based social information have attracted a lot of attentions. Although number social-based techniques been proposed in the literature, most their concepts are only individual or friends' rating behaviors. It leads to problem that recommended item list is usually constrained within users' living area. Furthermore, since context-aware and environmental changes quickly, especially networks, how select appropriate relevant users from such kind...

The research topic on transfer learning task has attracted a lot of attentions in recent years due to the wide applications. Although number techniques have been developed, basically they were designed manner and transferring among multiple source domains it was assumed that target domain share same feature space. However, with high variety issue under big data environments, this assumption violates scenario many real-world applications like activity recognition. In paper, we propose novel...