Data Mining Practical Machine Learning Tools and Techniques

527 articles related to imbalanced data and rare events are reviewed.Viewing reviewed papers from both technical and practical perspectives.Summarizing existing methods and corresponding statistics by a new taxonomy idea.Categorizing 162 application papers into 13 domains and giving introduction.Some opening questions are discussed at the end of this manuscript. Rare events, especially those that could potentially negatively impact society, often require humans decision-making responses. Detecting rare events can be viewed as a prediction task in data mining and machine learning communities. As these events are rarely observed in daily life, the prediction task suffers from a lack of balanced data. In this paper, we provide an in depth review of rare event detection from an imbalanced learning perspective. Five hundred and seventeen related papers that have been published in the past decade were collected for the study. The initial statistics suggested that rare events detection and imbalanced learning are concerned across a wide range of research areas from management science to engineering. We reviewed all collected papers from both a technical and a practical point of view. Modeling methods discussed include techniques such as data preprocessing, classification algorithms and model evaluation. For applications, we first provide a comprehensive taxonomy of the existing application domains of imbalanced learning, and then we detail the applications for each category. Finally, some suggestions from the reviewed papers are incorporated with our experiences and judgments to offer further research directions for the imbalanced learning and rare event detection fields.

Learning from class-imbalanced data

Evidence on foodborne disease (FBD) in low and middle income countries (LMICs) is still limited, but important studies in recent years have broadened our understanding. These suggest that developing country consumers are concerned about FBD; that most of the known burden of FBD disease comes from biological hazards; and, that most FBD is the result of consumption of fresh, perishable foods sold in informal markets. FBD is likely to increase in LMICs as the result of massive increases in the consumption of risky foods (livestock and fish products and produce) and lengthening and broadening value chains. Although intensification of agricultural production is a strong trend, so far agro-industrial production and modern retail have not demonstrated clear advantages in food safety and disease control. There is limited evidence on effective, sustainable and scalable interventions to improve food safety in domestic markets. Training farmers on input use and good practices often benefits those farmers trained, but has not been scalable or sustainable, except where good practices are linked to eligibility for export. Training informal value chain actors who receive business benefits from being trained has been more successful. New technologies, growing public concern and increased emphasis on food system governance can also improve food safety.

Food safety in low- and middle-income countries: What works, what doesn't and why

Effects of environmental conditions (temperature, pH, and glucose) on biofilm formation of Salmonella enterica serotype Kentucky and virulence gene expression.

Small molecule HDAC inhibitors: Promising agents for breast cancer treatment.

Histone deacetylases (HDAC) are emerging as promising targets in cancer, neuronal diseases and immune disorders. Computational modelling approaches have been widely applied for the virtual screening and rational design of novel HDAC inhibitors. In this study, different machine learning (ML) techniques were applied for the development of models that accurately discriminate HDAC2 inhibitors form non-inhibitors. The obtained models showed encouraging results, with the global accuracy in the external set ranging from 0.83 to 0.90. Various aspects related to the comparison of modelling techniques, applicability domain and descriptor interpretations were discussed. Finally, consensus predictions of these models were used for screening HDAC2 inhibitors from four chemical libraries whose bioactivities against HDAC1, HDAC3, HDAC6 and HDAC8 have been known. According to the results of virtual screening assays, structures of some hits with pair-isoform-selective activity (between HDAC2 and other HDACs) were revealed. This study illustrates the power of ML-based QSAR approaches for the screening and discovery of potent, isoform-selective HDACIs.

Quantitative structure–activity relationship analysis and virtual screening studies for identifying HDAC2 inhibitors from known HDAC bioactive chemical libraries

Pork is the most commonly consumed meat in Vietnam, and Salmonella enterica is a common contaminant. This study aimed to assess potential S. enterica cross-contamination between raw and cooked pork in Vietnamese households. Different scenarios for cross-contamination were constructed based on a household survey of pork handling practices (416 households). Overall, 71% of people used the same knife and cutting board for both raw and cooked pork; however, all washed their hands and utensils between handling raw and cooked pork. The different scenarios were experimentally tested. First, S. enterica was inoculated on raw pork and surfaces (hands, knives and cutting boards); next, water used for washing and pork were sampled to identify the presence and concentration of S. enterica during different scenarios of food preparation. Bootstrapping techniques were applied to simulate transfer rates of S. enterica cross-contamination. No cross-contamination to cooked pork was observed in the scenario of using the same hands with new cutting boards and knives. The probability of re-contamination in the scenarios involving re-using the cutting board after washing was significantly higher compared to the scenarios which used a new cutting board. Stochastic simulation found a high risk of cross-contamination from raw to cooked pork when the same hands, knives and cutting boards were used for handling raw and cooked pork (78%); when the same cutting board but a different knife was used, cross-contamination was still high (67%). Cross-contamination between was not seen when different cutting boards and knives were used for cutting raw and cooked pork. This study provided an insight into cross-contamination of S. enterica, given common food handling practices in Vietnamese households and can be used for risk assessment of pork consumption.

Simulating Cross-Contamination of Cooked Pork with Salmonella enterica from Raw Pork through Home Kitchen Preparation in Vietnam.

In many absorption, distribution, metabolism, and excretion (ADME) modeling problems, imbalanced data could negatively affect classification performance of machine learning algorithms. Solutions for handling imbalanced dataset have been proposed, but their application for ADME modeling tasks is underexplored. In this paper, various strategies including cost-sensitive learning and resampling methods were studied to tackle the moderate imbalance problem of a large Caco-2 cell permeability database. Simple physicochemical molecular descriptors were utilized for data modeling. Support vector machine classifiers were constructed and compared using multiple comparison tests. Results showed that the models developed on the basis of resampling strategies displayed better performance than the cost-sensitive classification models, especially in the case of oversampling data where misclassification rates for minority class have values of 0.11 and 0.14 for training and test set, respectively. A consensus model with enhanced applicability domain was subsequently constructed and showed improved performance. This model was used to predict a set of randomly selected high-permeability reference drugs according to the biopharmaceutics classification system. Overall, this study provides a comparison of numerous rebalancing strategies and displays the effectiveness of oversampling methods to deal with imbalanced permeability data problems.

Nam Nguyen-Hai

Papers

Quantitative structure–activity relationship analysis and virtual screening studies for identifying HDAC2 inhibitors from known HDAC bioactive chemical libraries

Simulating Cross-Contamination of Cooked Pork with Salmonella enterica from Raw Pork through Home Kitchen Preparation in Vietnam.

Exploring different strategies for imbalanced ADME data problem: case study on Caco-2 permeability modeling.

Multi-Criteria Decision Making: the Best Choice for the Modeling of Chemicals against Hyper-Pigmentation?