Randall Wald

Deep learning applications and challenges in big data analytics

Big Data Analytics and Deep Learning are two high-focus of data science. Big Data has become important as many organizations both public and private have been collecting massive amounts of domain-specific information, which can contain useful information about problems such as national intelligence, cyber security, fraud detection, marketing, and medical informatics. Companies such as Google and Microsoft are analyzing large volumes of data for business analysis and decisions, impacting existing and future technology. Deep Learning algorithms extract high-level, complex abstractions as data representations through a hierarchical learning process. Complex abstractions are learnt at a given level based on relatively simpler abstractions formulated in the preceding level in the hierarchy. A key benefit of Deep Learning is the analysis and learning of massive amounts of unsupervised data, making it a valuable tool for Big Data Analytics where raw data is largely unlabeled and un-categorized. In the present study, we explore how Deep Learning can be utilized for addressing some important problems in Big Data Analytics, including extracting complex patterns from massive volumes of data, semantic indexing, data tagging, fast information retrieval, and simplifying discriminative tasks. We also investigate some aspects of Deep Learning research that need further exploration to incorporate specific challenges introduced by Big Data Analytics, including streaming data, high-dimensional data, scalability of models, and distributed computing. We conclude by presenting insights into relevant future works by posing some questions, including defining data sampling criteria, domain adaptation modeling, defining criteria for obtaining useful data abstractions, improving semantic indexing, semi-supervised learning, and active learning.

A review of data mining using big data in health informatics

The amount of data produced within Health Informatics has grown to be quite vast, and analysis of this Big Data grants potentially limitless possibilities for knowledge to be gained. In addition, this information can improve the quality of healthcare offered to patients. However, there are a number of issues that arise when dealing with these vast quantities of data, especially how to analyze this data in a reliable manner. The basic goal of Health Informatics is to take in real world medical data from all levels of human existence to help advance our understanding of medicine and medical practice. This paper will present recent research using Big Data tools and approaches for the analysis of Health Informatics data gathered at multiple levels, including the molecular, tissue, patient, and population levels. In addition to gathering data at multiple levels, multiple levels of questions are addressed: human-scale biology, clinical-scale, and epidemic-scale. We will also analyze and examine possible future work for each of these areas, as well as how combining data from each level may provide the most promising approach to gain the most knowledge in Health Informatics.

Intrusion detection and Big Heterogeneous Data: a Survey

Intrusion Detection has been heavily studied in both industry and academia, but cybersecurity analysts still desire much more alert accuracy and overall threat analysis in order to secure their systems within cyberspace. Improvements to Intrusion Detection could be achieved by embracing a more comprehensive approach in monitoring security events from many different heterogeneous sources. Correlating security events from heterogeneous sources can grant a more holistic view and greater situational awareness of cyber threats. One problem with this approach is that currently, even a single event source (e.g., network traffic) can experience Big Data challenges when considered alone. Attempts to use more heterogeneous data sources pose an even greater Big Data challenge. Big Data technologies for Intrusion Detection can help solve these Big Heterogeneous Data challenges. In this paper, we review the scope of works considering the problem of heterogeneous data and in particular Big Heterogeneous Data. We discuss the specific issues of Data Fusion, Heterogeneous Intrusion Detection Architectures, and Security Information and Event Management (SIEM) systems, as well as presenting areas where more research opportunities exist. Overall, both cyber threat analysis and cyber intelligence could be enhanced by correlating security events across many diverse heterogeneous sources.

Feature Selection with High-Dimensional Imbalanced Data

Feature selection is an important topic in data mining, especially for high dimensional datasets. Filtering techniques in particular have received much attention, but detailed comparisons of their performance is lacking. This work considers three filters using classifier performance metrics and six commonly-used filters. All nine filtering techniques are compared and contrasted using five different microarray expression datasets. In addition, given that these datasets exhibit an imbalance between the number of positive and negative examples, the utilization of sampling techniques in the context of feature selection is examined.

A review of the stability of feature selection techniques for bioinformatics data

Feature selection is an important step in data mining and is used in various domains including genetics, medicine, and bioinformatics. Choosing the important features (genes) is essential for the discovery of new knowledge hidden within the genetic code as well as the identification of important biomarkers. Although feature selection methods can help sort through large numbers of genes based on their relevance to the problem at hand, the results generated tend to be unstable and thus cannot be reproduced in other experiments. Relatedly, research interest in the stability of feature ranking methods has grown recently and researchers have produced experimental designs for testing the stability of feature selection, creating new metrics for measuring stability and new techniques designed to improve the stability of the feature selection process. In this paper, we will introduce the role of stability in feature selection with DNA microarray data. We list various ways of improving feature ranking stability, and discuss feature selection techniques, specifically explaining ensemble feature ranking and presenting various ensemble feature ranking aggregation methods. Finally, we discuss experimental procedures such as dataset perturbation, fixed overlap partitioning, and cross validation procedures that help researchers analyze and measure the stability of feature ranking methods. Throughout this work, we investigate current research in the field and discuss possible avenues of continuing such research efforts.

Comparative Analysis of DNA Microarray Data through the Use of Feature Selection Techniques

One of today’s most important scientific research topics is discovering the genetic links between cancers. This paper contains the results of a comparison of three different cancers (breast, colon, and lung) based on the results of feature selection techniques on a data set created from DNA micro array data consisting of samples from all three cancers. The data was run through a set of eighteen feature rankers which ordered the genes by importance with respect to a targeted cancer. This process was repeated three times, each time with a different target cancer. The rankings were then compared, keeping each feature ranker static while varying the cancers being compared. The cancers were evaluated both in pairs and all together, for matching genes. The results of the comparison show a large correlation between the two known hereditary cancers, breast and colon, and little correlation between lung cancer and the other cancers. This is the first study to apply eighteen different feature rankers in a bioinformatics case study, eleven of which were recently proposed and implemented by our research team.

Random forest: A reliable tool for patient response prediction

The goal of classification is to reliably identify instances that are members of the class of interest. This is especially important for predicting patient response to drugs. However, with high dimensional datasets, classification is both complicated and enhanced by the feature selection process. When designing a classification experiment there are a number of decisions which need to be made in order to maximize performance. These decisions are especially difficult for researchers in fields where data mining is not the focus, such as patient response prediction. It would be easier for such researchers to make these decisions if either their outcomes were chosen or their scope reduced, by using a learner which minimizes the impact of these decisions. We propose that Random Forest, a popular ensemble learner, can serve this role. We performed an experiment involving nineteen different feature selection rankers (eleven of which were proposed and implemented by our research team) to thoroughly test both the Random Forest learner and five other learners. Our research shows that, as long as a large enough number of features are used, the results of using Random Forest are favorable regardless of the choice of feature selection strategy, showing that Random Forest is a suitable choice for patient response prediction researchers who want to do not wish to choose from amongst a myriad of feature selection approaches.

Using Twitter Content to Predict Psychopathy

An ever-growing number of users share their thoughts and experiences using the Twitter micro logging service. Although sometimes dismissed as containing too little content to convey significant information, these messages can be combined to build a larger picture of the user posting them. One particularly notable personality trait which can be discovered this way is psychopathy: the tendency for disregarding others and the rule of society. In this paper, we explore techniques to apply data mining towards the goal of identifying those who score in the top 1.4% of a well-known psychopathy metric using information available from their Twitter accounts. We apply a newly-proposed form of ensemble learning, Select RUSBoost (which adds feature selection to our earlier imbalance-aware ensemble in order to resolve high dimensionality), employ four classification learners, and use four feature selection techniques. The results show that when using the optimal choices of techniques, we are able to achieve an AUC value of 0.736. Furthermore, these results were only achieved when using the Select RUSBoost technique, demonstrating the importance of feature selection, data sampling, and ensemble learning. Overall, we show that data mining can be a valuable tool for law enforcement and others interested in identifying abnormal psychiatric states from Twitter data.

A comparative study of iterative and non-iterative feature selection techniques for software defect prediction

Two important problems which can affect the performance of classification models are high-dimensionality (an overabundance of independent features in the dataset) and imbalanced data (a skewed class distribution which creates at least one class with many fewer instances than other classes). To resolve these problems concurrently, we propose an iterative feature selection approach, which repeated applies data sampling (in order to address class imbalance) followed by feature selection (in order to address high-dimensionality), and finally we perform an aggregation step which combines the ranked feature lists from the separate iterations of sampling. This approach is designed to find a ranked feature list which is particularly effective on the more balanced dataset resulting from sampling while minimizing the risk of losing data through the sampling step and missing important features. To demonstrate this technique, we employ 18 different feature selection algorithms and Random Undersampling with two post-sampling class distributions. We also investigate the use of sampling and feature selection without the iterative step (e.g., using the ranked list from a single iteration, rather than combining the lists from multiple iterations), and compare these results from the version which uses iteration. Our study is carried out using three groups of datasets with different levels of class balance, all of which were collected from a real-world software system. All of our experiments use four different learners and one feature subset size. We find that our proposed iterative feature selection approach outperforms the non-iterative approach.

Threshold-based feature selection techniques for high-dimensional bioinformatics data

Analysis conducted for bioinformatics applications often requires the use of feature selection methodologies to handle datasets with very high dimensionality. We propose 11 new threshold-based feature selection techniques and compare the performance of these new techniques to that of six standard filter-based feature selection procedures. Unlike other comparisons of feature selection techniques, we directly compare the feature rankings produced by each technique using Kendall’s Tau rank correlation, showing that the newly proposed techniques exhibit substantially different behaviors than the standard filter-based feature selection methods. Our experiments consider 17 different bioinformatics datasets, and the similarities of the feature selection techniques are analyzed using the Frobenius norm. The feature selection techniques are also compared by using Naive Bayes and Support Vector Machine algorithms to learn from the training datasets. The experimental results show that the new procedures perform very well compared to the standard filters, and hence are useful feature selection methodologies for the analysis of bioinformatics data.