Exploring the integration of data mining and data visualization

Due to the rapid advances in computing and sensing technologies, enormous amounts of data are being generated everyday in various applications. The integration of data mining and data visualization has been widely used to analyze these massive and complex data sets to discover hidden patterns. For both data mining and visualization to be effective, it is important to include the visualization techniques in the mining process and to generate the discovered patterns for a more comprehensive visual view. In this dissertation, four related problems: dimensionality reduction for visualizing high dimensional datasets, visualization-based clustering evaluation, interactive document mining, and multiple clusterings exploration are studied to explore the integration of data mining and data visualization. In particular, we 1) propose an efficient feature selection method (reliefF + mRMR) for preprocessing high dimensional datasets; 2) present DClusterE to integrate cluster validation with user interaction and provide rich visualization tools for users to examine document clustering results from multiple perspectives; 3) design two interactive document summarization systems to involve users efforts and generate customized summaries from 2D sentence layouts; and 4) propose a new framework which organizes the different input clusterings into a hierarchical tree structure and allows for interactive exploration of multiple clustering solutions.

Longitudinal effects of family factors on alcohol use and alcohol-related problems among males from pre-adolescence to adulthood

The goal of this study was to examine the longitudinal effects of five family factors on alcohol use among adolescent males. The family factors included familism (family pride, loyalty, and cohesion), parent derogation (being put down by parents), parent/child communication, family alcohol problems and family drug problems. The study focused on the effects of the family factors reported by a sample of 451 White-non-Hispanic and African American males during early and mid-adolescence on (1) the intensity of alcohol use in mid-adolescence, and (2) the number of problems associated with alcohol use during the transition to young adulthood. The study also explored racial differences in the effects of the family factors. The data for this study were derived from a two-phase longitudinal epidemiologic cohort study of male and female adolescents enrolled in middle schools in Miami, FL. Data were collected at four points between 1990 and 2001. Linear and logistical regressions were used to analyze the effects of the family variables on the dependent variables. ^ The results of the analyses indicated that all of the family variables except family drug problems were statistically significant predictors of the level of alcohol use and alcohol-related problems. Familism had a moderate influence on both of the dependent variables at all data points, while parent derogation, parent/child communication and family alcohol problems were weak predictors. While the family factors varied by race, their impact on the dependent variables did not vary substantially. ^ This study had methodological shortcomings related to measurement and design that may have contributed to the weak influence of the variables. Future studies should explore possible mediating effects of these variables, and should employ more sensitive measures that are culturally appropriate. The results suggest that, since early family factors have long-term effects on children's substance-using behaviors, the family environment should be addressed in prevention and intervention efforts. ^

Impact of a comprehensive nutrition and lifestyle education intervention on body weight and health-related outcomes in morbidly-obese Hispanic-Americans following laparoscopic Roux-en-Y gastric bypass

As morbid obesity increasingly affects Hispanic-Americans, the incidence of Roux-en-Y gastric bypass procedures (RYGB) among this population rises. Prospective research on the impact of postoperative educational interventions focused on Hispanic- Americans is needed to prevent premature weight loss plateau, weight regain, nutritional deficiencies, and relapse of obesity-related comorbidities. This randomized-controlled study evaluated the impact of a comprehensive nutrition and lifestyle education intervention (6 biweekly postoperative sessions that incorporated motivational strategies for behavioral change) as compared to a non-comprehensive approach (printed guidelines for healthy lifestyle). The variables to consider are body weight, obesity-related comorbidities (depression, diabetes, dyslipidemia, and others), nutrient status, physical activity, and eating habits in 144 morbidly-obese adult Hispanic-Americans 6 to 12 months following RYGB. Patients were randomly assigned to either the comprehensive intervention (n=72) or the comparison group (n=72). Participants (mean age 44.5 ± 13.5 years) were mainly Cuban-born females (83.3%). Intervention sessions attendance was 64%. At 12 months, both groups lost weight significantly, but those in the comprehensive intervention experienced greater excess weight loss than those in the comparison group (80% vs. 64% from preoperative excess weight, P<.001). Intervention participants were significantly more involved in physical activity (+ 14 min/week vs. – 4 min/week), had decreased depression, joint illness, and required less medication for comorbidities than comparison participants. Additionally, those in the comprehensive intervention had sustained supplement intake experiencing less folate deficiency (P=.014). The non-comprehensive intervention group significantly decreased their protein and supplement intake compared to the intervention group. Patients in the comprehensive intervention had significantly better eating habits reflected by fewer episodes of dumping syndrome, constipation, and night eating, than those in the comparison group who reported greater eating in response to negative emotions (P=.003). These findings support the importance of a comprehensive educational approach to achieve more effective weight reduction and health-related outcomes to prevent relapse of obesity-related comorbidities and nutritional deficiencies in Hispanic-Americans 6 to 12 months following RYGB.

Document understanding using data mining and machine learning techniques

With the explosive growth of the volume and complexity of document data (e.g., news, blogs, web pages), it has become a necessity to semantically understand documents and deliver meaningful information to users. Areas dealing with these problems are crossing data mining, information retrieval, and machine learning. For example, document clustering and summarization are two fundamental techniques for understanding document data and have attracted much attention in recent years. Given a collection of documents, document clustering aims to partition them into different groups to provide efficient document browsing and navigation mechanisms. One unrevealed area in document clustering is that how to generate meaningful interpretation for the each document cluster resulted from the clustering process. Document summarization is another effective technique for document understanding, which generates a summary by selecting sentences that deliver the major or topic-relevant information in the original documents. How to improve the automatic summarization performance and apply it to newly emerging problems are two valuable research directions. To assist people to capture the semantics of documents effectively and efficiently, the dissertation focuses on developing effective data mining and machine learning algorithms and systems for (1) integrating document clustering and summarization to obtain meaningful document clusters with summarized interpretation, (2) improving document summarization performance and building document understanding systems to solve real-world applications, and (3) summarizing the differences and evolution of multiple document sources.

The 4-aza-S-ribosyl-L-homocysteine derivatives and the related gamma-lactam and azahemiacetal analogs: Synthesis, inhibition and quorum sensing activity

Quorum sensing (QS) is a population-dependent signaling process bacteria use to control multiple processes including virulence, critical for establishing infection. There are two major pathways of QS systems. Type 1 is species specific or intra-species communication in which N-acylhomoserine lactones (Gram-negative bacteria) or oligopeptides (Gram-positive bacteria) are employed as signaling molecules (autoinducer one). Type 2 is inter-species communication in which S-4,5-dihydroxy-2,3-pentanedione (DPD) or its borate esters are used as signaling molecules. The DPD is biosynthesized by LuxS enzyme from S-ribosylhomocysteine (SRH). Recent increase in prevalence of bacterial strains resistant to antibiotics emphasizes the need for the development of new generation of antibacterial agents. Interruption of QS by small molecules is one of the viable options as it does not affect bacterial growth but only virulence, leading to less incidence of microbial resistance. Thus, in this work, inhibitors of both N-acylhomoserine lactone (AHL) mediated intra-species and LuxS enzyme, involved in inter-species QS are targeted. The γ-lactam and their reduced cyclic azahemiacetal analogs, bearing the additional alkylthiomethyl substituent, were designed and synthesized targeting AHL mediated QS systems in P. aeruginosa and Vibrio harveyi. The γ-lactams with nonylthio or dodecylthio chains acted as inhibitors of las signaling in P. aeruginosa with moderate potency. The cyclic azahemiacetal with shorter propylthio or hexylthio substituent were found to strongly inhibit both las and rhl signaling in P. aeruginosa at higher concentrations. However, lactam and their azahemiacetal analogs were found to be inactive in V. harveyi QS systems. The 4-aza-S-ribosyl-L-homocysteine (4-aza-SRH) analogs and 2-deoxy-2-substituted-S-ribosyl-L-homocysteine analogs were designed and synthesized targeting Bacillus subtilis LuxS enzyme. The 4-aza-SRH analogs in which oxygen in ribose ring is replaced by nitrogen were further modified at anomeric position to produce pyrrolidine, lactam, nitrone, imine and hemiaminal analogs. Pyrrolidine and lactam analogs which lack anomeric hydroxyl, acted as competitive inhibitors of LuxS enzyme with KI value of 49 and 37 µM respectively. The 2,3-dideoxy lactam analogs were devoid of activity. Such findings attested the significance of hydroxyl groups for LuxS binding and activity. Hemiaminal analog of SRH was found to be a time-dependent inhibitor with IC50 value of 60 µM.

Correlates of Boater Knowledge and Views Regarding the Florida Manatee and Manatee Related Issues

Scientists are extremely concerned that the Florida manatee, Trichechus manatus latirostris, does not have a sufficient population number nor reproductive rate to ensure long-term species viability. This research was done to investigate aspects of boater behavior in an effort to determine measures that might decrease boat-manatee collisions and help researchers plan intervention strategies in related areas. Initially, boat data was collected over a period of several months at a Miami waterway. Next, a detailed phone survey of boat owners was conducted. Although most boaters were found noncompliant with the speed zone, several questions from the survey showed that most valued manatees and almost half wanted them to remain listed as endangered. This is very encouraging as boater support for the manatee is possibly one of the most important indicators of manatee chances for survival. Suggestions included requiring mandatory boating classes and informing boaters of faster routes to their destinations.

Physical activity and nutrition related variables in hispanic and non-hispanic college students

It has been estimated that one in four adults have sedentary lifestyles. In addition there appears to be an increase in obesity across the life span. It is of great importance to the health of this nation to understand how to promote more active lifestyles through the identification of lifestyle behaviors of active individuals and potential predictors of physical activity (PA). Seven hundred and seventy-seven college students were surveyed to investigate the relationship between nutrition related variables (i.e., dietary restraint, nutrition knowledge, food choice and body weight concerns) and PA. In this study, over half of the students reported doing 30 minutes of moderate intensity PA daily. Vigorously active males and females chose low fat foods more often than the less active group. Exercisers and non-exercisers had similar nutrition knowledge. The results of this study suggest that students who are more active are more conscience about making healthier food choices.

Mining the online social network data: Influence, summarization, and organization

Online Social Network (OSN) services provided by Internet companies bring people together to chat, share the information, and enjoy the information. Meanwhile, huge amounts of data are generated by those services (they can be regarded as the social media ) every day, every hour, even every minute, and every second. Currently, researchers are interested in analyzing the OSN data, extracting interesting patterns from it, and applying those patterns to real-world applications. However, due to the large-scale property of the OSN data, it is difficult to effectively analyze it. This dissertation focuses on applying data mining and information retrieval techniques to mine two key components in the social media data — users and user-generated contents. Specifically, it aims at addressing three problems related to the social media users and contents: (1) how does one organize the users and the contents? (2) how does one summarize the textual contents so that users do not have to go over every post to capture the general idea? (3) how does one identify the influential users in the social media to benefit other applications, e.g., Marketing Campaign? The contribution of this dissertation is briefly summarized as follows. (1) It provides a comprehensive and versatile data mining framework to analyze the users and user-generated contents from the social media. (2) It designs a hierarchical co-clustering algorithm to organize the users and contents. (3) It proposes multi-document summarization methods to extract core information from the social network contents. (4) It introduces three important dimensions of social influence, and a dynamic influence model for identifying influential users.

Mining the Online Social Network Data: Influence, Summarization, and Organization

Online Social Network (OSN) services provided by Internet companies bring people together to chat, share the information, and enjoy the information. Meanwhile, huge amounts of data are generated by those services (they can be regarded as the social media ) every day, every hour, even every minute, and every second. Currently, researchers are interested in analyzing the OSN data, extracting interesting patterns from it, and applying those patterns to real-world applications. However, due to the large-scale property of the OSN data, it is difficult to effectively analyze it. This dissertation focuses on applying data mining and information retrieval techniques to mine two key components in the social media data — users and user-generated contents. Specifically, it aims at addressing three problems related to the social media users and contents: (1) how does one organize the users and the contents? (2) how does one summarize the textual contents so that users do not have to go over every post to capture the general idea? (3) how does one identify the influential users in the social media to benefit other applications, e.g., Marketing Campaign? The contribution of this dissertation is briefly summarized as follows. (1) It provides a comprehensive and versatile data mining framework to analyze the users and user-generated contents from the social media. (2) It designs a hierarchical co-clustering algorithm to organize the users and contents. (3) It proposes multi-document summarization methods to extract core information from the social network contents. (4) It introduces three important dimensions of social influence, and a dynamic influence model for identifying influential users.

Academic performance of college students in financial management and related prerequisite courses: A comparison of senior high school and commercial high school graduates attending Tamsui Oxford University College in Taiwan

In Taiwan, the college freshmen are recruited graduates of both senior high school and senior vocational school. The Ministry of Education (MOE) of the Republic of China prescribes the standards of curriculum and equipment for schools at all levels and categories. There exists a considerably different curriculum arrangement for senior high schools and vocational high schools in Taiwan at the present time. The present study used a causal-comparative research design to identify the influences of different post-secondary educational background on specialized course performance of college business majors. ^ The students involved in this study were limited to the students of four business-related departments at Tamsui Oxford University College in Taiwan. Students were assigned to comparison groups based on their post-secondary educational background as senior high school graduates and commercial high school graduates. The analysis of this study included a comparison of students' performance on lower level courses and a comparison of students' performance in financial management. The analysis also considered the relationship between the students' performance in financial management and its related prerequisite courses. The Kolb Learning Style Inventory (LSI) survey was administered to categorize subjects' learning styles and to compare the learning styles between the two groups in this study. The applied statistical methods included t-test, correlation, multiple regression, and Chi-square. ^ The findings of this study indicated that there were significant differences between the commercial high school graduates and the senior high school graduates on academic performances in specialized courses but not in general courses. There were no significant differences in learning styles between the two groups. These findings lead to the conclusion that business majors' academic performance in specialized courses were influenced by their post-secondary educational background. ^

Mining in Peru: Indigenous and Peasant Communities vs. The State and Mining Capital

The Peruvian economy depends for its growth on the export of natural resources and investment in the mining and hydrocarbon sectors. Peruvian governments and mining corporations have confronted anti-mining protests in different ways. While the current government has introduced policies of social inclusion to soften the negative effects of the operations of mining capital and policies of dialogue to engage social actors with the essence of governmental policies, mining companies use corporate social responsibility programs as a cover for the devastating effects of their operations on the environment and the livelihoods and habitats of the indigenous and peasant communities. Curiously, in the current context of the declining commodity prices and export volumes the Peruvian government strengthens its extractivist model of development. This article argues that whatever government that follows the rules of capital cannot but favor the corporations. It points out the main adversaries of the indigenous and peasant communities and the problems to transform the locally and/or regionally struggle into a nationwide battle for another development model.

Process inventory and pollution prevention overview for the citrus industry

Florida citrus represents approximately 70 percent of the industry production in the United States; therefore, any associated agricultural and industrial contamination is of concern and a focus of attention. The use of synthetic organic chemicals has become a farmer's necessity in order to supply consumers with high quality products, free of pest damage. However, industrial citrus wastes and chemical residual levels worry not only government agencies but also consumers since they indicate a serious habitat risk. This study assesses citrus industrial processes and the paths that chemical substances follow from the time the citrus seed is planted until consumers get a final product as either fresh fruit or processed product. The study is built on information from United States Environmental Protection Agency (US EPA) manuals, Dade County Environmental Resources Management (DERM) inspection records, United States Food and Drug Administration (US FDA) regulations, Florida standards, journal publications, and research reports. Pollution prevention (P2 or prevention-of-pollution) alternatives are identified; alternatives are proposed, evaluated, and included. Strategies are described and pollution prevention opportunities proposed to minimize citrus wastes generation, chemical residuals in products, their environmental impact and health risk aspects while maximizing product quality.