Yu-Feng Lin

Mining high utility episodes in complex event sequences

Frequent episode mining (FEM) is an interesting research topic in data mining with wide range of applications. However, the traditional framework of FEM treats all events as having the same importance/utility and assumes that a same type of event appears at most once at any time point. These simplifying assumptions do not reflect the characteristics of scenarios in real applications and thus the useful information of episodes in terms of utilities such as profits is lost. Furthermore, most studies on FEM focused on mining episodes in simple event sequences and few considered the scenario of complex event sequences, where different events can occur simultaneously. To address these issues, in this paper, we incorporate the concept of utility into episode mining and address a new problem of mining high utility episodes from complex event sequences, which has not been explored so far. In the proposed framework, the importance/utility of different events is considered and multiple events can appear simultaneously. Several novel features are incorporated into the proposed framework to resolve the challenges raised by this new problem, such as the absence of anti-monotone property and the huge set of candidate episodes. Moreover, an efficient algorithm named UP-Span (Utility ePisodes mining by Spanning prefixes) is proposed for mining high utility episodes with several strategies incorporated for pruning the search space to achieve high efficiency. Experimental results on real and synthetic datasets show that UP-Span has excellent performance and serves as an effective solution to the new problem of mining high utility episodes from complex event sequences.

Reliable Early Classification on Multivariate Time Series with Numerical and Categorical Attributes

Early classification on multivariate time series has recently emerged as a novel and important topic in data mining fields with wide applications such as early detection of diseases in healthcare domains. Most of the existing studies on this topic focused only on univariate time series, while some very recent works exploring multivariate time series considered only numerical attributes and are not applicable to multivariate time series containing both of numerical and categorical attributes. In this paper, we present a novel methodology named REACT (Reliable EArly ClassificaTion), which is the first work addressing the issue of constructing an effective classifier on multivariate time series with numerical and categorical attributes in serial manner so as to guarantee stability of accuracy compared to the classifiers using full-length time series. Furthermore, we also employ the GPU parallel computing technique to develop an extended mechanism for building the early classifier efficiently. Experimental results on real datasets show that REACT significantly outperforms the state-of-the-art method in terms of accuracy and earliness, and the GPU implementation is verified to substantially enhance the efficiency by several orders of magnitudes.

Discovering utility-based episode rules in complex event sequences

Abstract Mining high utility episode rules in complex event sequences has emerged as an important topic in data mining because the utility-based episode rules generated may provide important insights that facilitate decision making for expert and intelligent systems. Although one may employ previous methods in this research area to indirectly construct utility-based episode rules, they typically lack efficiency and effectiveness for real-world applications. In this paper, we develop a novel methodology to directly generate high utility episode rules during the mining process, which is the first work addressing the issue of utility-based episode rule mining. Our goal is to simultaneously resolve the difficulty of the previous reported methods for frequent episode mining and utility-based episode mining. An algorithm called UBER-Mine (Utility-Based Episode Rules) and a structure named UR-Tree (Utility Rule Tree) are proposed to mine efficiently the complete set of high utility episode rules in complex event sequences. In short, UBER-Mine is based on an extended downward closure property, which can efficiently discover utility-based episode rules. On the other hand, UR-Tree can maintain important event information without producing candidate episodes to further accelerate the mining process. Results on both real and synthetic datasets show that UBER-Mine with UR-Tree has good scalability on large datasets and runs faster than the basic UBER-Mine and the current best high utility episode mining algorithm over 100 times. Furthermore, by proposing a high-utility episode-rule model called IV-UBER (InVestment by Utility-Based Episode Rules), we further demonstrate the effectiveness of our method for mining high utility-based episode rules on a real-world application for stock investment. The experimental results show that our proposed IV-UBER method outperforms several state-of-the-art algorithms in terms of both precision and annualized return for investment.

A Novel Episode Mining Methodology for Stock Investment

In this paper, we present a novel methodology for stock investment using episode mining and technical indicators. The time-series data of stock price and the derived moving average, a class of well-known technical indicators, are used for the construction of complex episode events and rules. Our objective is to devise a profitable episode-based investment model to mine associated events in the stock market. Using Taiwan Capitalization Weighted Stock Index (TAIEX), the empirical results show that our proposed model significantly outperforms the benchmark in terms of cumulative total returns. We also show that the level of the precision by our model is close to 60%, which is better than random guessing. Based upon the results obtained, we expect this novel episode-based methodology will advance the research in data mining for computational finance and provide an alternative to stock investment in practice.

A Mobile Framework for Personalized Diabetes Telecare

The latest estimates for worldwide diabetic population is about 366 million, with the expected diabetic population in 2030 at 552 million. The steadily increasing diabetic population and strained health resources call for a more efficient approach to diabetes care. In the past few years, smart phones have become the mainstream communications device, especially for the younger generation. Coupled with the increased ease in developing and deploying mobile applications, there has been a surge in medical and health related mobile programs. These modern smart phones provide a readily available platform for deploying healthcare related services. By employing their computing resources and connectivity, health institutions can easily provide telehealth services to chronically ill patients. In this work, we put forward a mobile framework for providing personalized diabetes telehealth. Through leveraging the power of smart phones and wireless connectivity, better care and improved quality of life can be provided for diabetic patients by this framework.

Design of a Real-Time and Continua-Based Framework for Care Guideline Recommendations

Telehealth is an important issue in the medical and healthcare domains. Although a number of systems have been developed to meet the demands of emerging telehealth services, the following problems still remain to be addressed: (1) most systems do not monitor/predict the vital signs states so that they are able to send alarms to caregivers in real-time; (2) most systems do not focus on reducing the amount of work that caregivers need to do, and provide patients with remote care; and (3) most systems do not recommend guidelines for caregivers. This study thus proposes a framework for a real-time and Continua-based Care Guideline Recommendation System (Cagurs) which utilizes mobile device platforms to provide caregivers of chronic patients with real-time care guideline recommendations, and that enables vital signs data to be transmitted between different devices automatically, using the Continua standard. Moreover, the proposed system adopts the episode mining approach to monitor/predict anomalous conditions of patients, and then offers related recommended care guidelines to caregivers so that they can offer preventive care in a timely manner.

Mining Sequential Risk Patterns From Large-Scale Clinical Databases for Early Assessment of Chronic Diseases: A Case Study on Chronic Obstructive Pulmonary Disease

Chronic diseases have been among the major concerns in medical fields since they may cause a heavy burden on healthcare resources and disturb the quality of life. In this paper, we propose a novel framework for early assessment on chronic diseases by mining sequential risk patterns with time interval information from diagnostic clinical records using sequential rules mining, and classification modeling techniques. With a complete workflow, the proposed framework consists of four phases namely data preprocessing, risk pattern mining, classification modeling, and post analysis. For empiricasl evaluation, we demonstrate the effectiveness of our proposed framework with a case study on early assessment of COPD. Through experimental evaluation on a large-scale nationwide clinical database in Taiwan, our approach can not only derive rich sequential risk patterns but also extract novel patterns with valuable insights for further medical investigation such as discovering novel markers and better treatments. To the best of our knowledge, this is the first work addressing the issue of mining sequential risk patterns with time-intervals as well as classification models for early assessment of chronic diseases.

A novel methodology for stock investment using high utility episode mining and genetic algorithm

Abstract In this paper, we present a novel methodology for stock investment using the technique of high utility episode mining and genetic algorithms. Our objective is to devise a profitable episode-based investment model to reveal hidden events that are associated with high utility in the stock market. The time series data of stock price and the derived technical indicators, including moving average, moving average convergence and divergence, random index and bias index, are used for the construction of episode events. We then employ the genetic algorithm for the simultaneous optimization on parameters and selection of subsets of models. The empirical results show that our proposed method significantly outperforms the state-of-the-art methods in terms of annualized returns of investment and precision. We also provide a set of Z -tests to statistically validate the effectiveness of our proposed method. Based upon the promising results obtained, we expect this novel methodology can advance the research in data mining for computational finance and provide an alternative to stock investment in practice.

Mining disease sequential risk patterns from nationwide clinical databases for early assessment of chronic obstructive pulmonary disease

Chronic diseases may cause heavy burden on health care resources and disturb the quality of life. Chronic Obstructive Pulmonary Disease (COPD) is an important chronic disease, which takes a long period of time to progress and hard to detect in early stage. In this work, we propose a novel approach for early assessment on COPD by mining COPD-related sequential risk patterns from diagnostic clinical records using sequential rule mining and classification techniques. Through experimental evaluation on a large-scale nationwide clinical database in Taiwan, our approach is shown to be not only capable of deriving many sequential risk patterns, but also reliable in prediction results. Moreover, the discovered sequential risk patterns may provide potential clues for physicians to derive novel markers for early detection on COPD. To our best knowledge, this is the first work that addresses the important issue of early assessment on COPD through mining sequential risk patterns from large-scale clinical databases.

An interactive healthcare system with personalized diet and exercise guideline recommendation

Recently people pay more and more attention on how to effectively and efficiently analyze the result of regular physical examinations to provide the most helpful information for individual health management. In this paper, we design and develop an interactive system of virtual healthcare assistant to help people, especially for those who suffer from chronic diseases (e.g., metabolic syndrome) to easily understand their health conditions and then well manage it. This system analyzes the result of regular physical examination to evaluate the health risk and provide personalized healthcare services for users in terms of diet and exercise guideline recommendations. We developed some interactive ways for users to easily feedback their vital signs to the system and quickly get the suggestions for health management from the system. Besides the browser-based system, we also developed a mobile App that can regularly remind users to carry out the recommendations, which are provided by the system. To prove the system is feasible in the real-world clinical environment, we also applied the Institutional Review Board (IRB) for a human subject research to validate this system. Other than the functional features, there are also several important non-functional features of the extensibility and the convenience for use. First, we use the physical examination result as the raw data to be analyzed. Its very convenient for users with very low cost. Second, the system design is extendable, so it can be easily adjusted to work for any chronic ills, even other kinds of diseases. Moreover, it can be extended to provide other kinds of healthcare guideline recommendations as well. These features constitute the main contributions of this work.