scholarly journals Performance Evaluation of Pre-processing Techniques on Diabetes Prediction

2021 ◽  
Vol 6 (4) ◽  
Author(s):  
Aminat B. Yusuf ◽  
Ogar O. Austin ◽  
Shinaigo Y. Tadi ◽  
Fatsuma Jauro
Keyword(s):  
Random Forest ◽  
Missing Values ◽  
Nearest Neighbor ◽  
Pearson Correlation ◽  
Class Imbalance ◽  
K Nearest Neighbor ◽  
Improved Model ◽  
Percent Accuracy ◽  
Sampling Procedures ◽  
Processing Techniques

Medical industry contains a large amount of sensitive data that must be evaluated in order to get insight into records. The nonlinearity, non-normality, correlation structures and complicated diabetic medical records, on the other hand, makes accurate predictions difficult. The Pima Indian Diabetes dataset is one of them, owing to the dataset's imbalance, large number of missing values and difficulty in identifying highly risk factors. Some of these challenges have been solved using computational approaches such as machine learning methods, but they have not performed ideally, with pre-processing techniques being recognized as critical to achieving correct findings. The goal of this work is to apply multiple pre-processing approaches to increase the accuracy of some simple models. These multiple pre-processing techniques are median imputation in which null values are substituted by finding the median of the input variables dependent on whether or not the patient is diabetic and then follow by applying oversampling and under-sampling procedures on both majority and minority votes. These votes are applied in order to address the problem of class imbalance as pointed out from the literature. Finally, the dimension reduction Pearson correlation is used to detect high-risk features since it is effective at quantifying information between attributes and their labels. In this study, these techniques are applied in the same order to Linear Regression, Naive Bayes, Decision Tree, K Nearest Neighbor, Random Forest and Gaussian Boosting classifiers. The utility of the techniques on the mentioned classifiers is validated using performance measures such as Accuracy, Precision and Recall.  The Random Forest Classifier is found to be the best-improved model, with 95 percent accuracy, 94.25 percent precision and 95.35 percent recall. Medical practitioners may find the provided strategies beneficial in improving the efficiency of diabetes analysis. Keywords— Classifiers, diabetes, Pima Indian Diabetes dataset, pre-processing techniques

scholarly journals An in-depth performance analysis of the oversampling techniques for high-class imbalanced dataset

2021 ◽  
Vol 7 (1) ◽  
pp. 63
Author(s):  
Prasetyo Wibowo ◽  
Chastine Fatichah
Keyword(s):  
Random Forest ◽  
Performance Analysis ◽  
Nearest Neighbor ◽  
Class Imbalance ◽  
K Nearest Neighbor ◽  
Class Imbalance Problem ◽  
Minority Class ◽  
High Class ◽  
Imbalance Problem ◽  
Imbalanced Classes

Class imbalance occurs when the distribution of classes between the majority and the minority classes is not the same. The data on imbalanced classes may vary from mild to severe. The effect of high-class imbalance may affect the overall classification accuracy since the model is most likely to predict most of the data that fall within the majority class.  Such a model will give biased results, and the performance predictions for the minority class often have no impact on the model. The use of the oversampling technique is one way to deal with high-class imbalance, but only a few are used to solve data imbalance. This study aims for an in-depth performance analysis of the oversampling techniques to address the high-class imbalance problem. The addition of the oversampling technique will balance each class’s data to provide unbiased evaluation results in modeling. We compared the performance of Random Oversampling (ROS), ADASYN, SMOTE, and Borderline-SMOTE techniques. All oversampling techniques will be combined with machine learning methods such as Random Forest, Logistic Regression, and k-Nearest Neighbor (KNN). The test results show that Random Forest with Borderline-SMOTE gives the best value with an accuracy value of 0.9997, 0.9474 precision, 0.8571 recall, 0.9000 F1-score, 0.9388 ROC-AUC, and 0.8581 PRAUC of the overall oversampling technique.

scholarly journals Evaluation of Three Different Machine Learning Methods for Object-Based Artificial Terrace Mapping—A Case Study of the Loess Plateau, China

2021 ◽  
Vol 13 (5) ◽  
pp. 1021
Author(s):  
Hu Ding ◽  
Jiaming Na ◽  
Shangjing Jiang ◽  
Jie Zhu ◽  
Kai Liu ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Loess Plateau ◽  
Water Conservation ◽  
Nearest Neighbor ◽  
Gradient Boosting ◽  
K Nearest Neighbor ◽  
The Loess Plateau ◽  
Object Based ◽  
Extreme Gradient Boosting

Artificial terraces are of great importance for agricultural production and soil and water conservation. Automatic high-accuracy mapping of artificial terraces is the basis of monitoring and related studies. Previous research achieved artificial terrace mapping based on high-resolution digital elevation models (DEMs) or imagery. As a result of the importance of the contextual information for terrace mapping, object-based image analysis (OBIA) combined with machine learning (ML) technologies are widely used. However, the selection of an appropriate classifier is of great importance for the terrace mapping task. In this study, the performance of an integrated framework using OBIA and ML for terrace mapping was tested. A catchment, Zhifanggou, in the Loess Plateau, China, was used as the study area. First, optimized image segmentation was conducted. Then, features from the DEMs and imagery were extracted, and the correlations between the features were analyzed and ranked for classification. Finally, three different commonly-used ML classifiers, namely, extreme gradient boosting (XGBoost), random forest (RF), and k-nearest neighbor (KNN), were used for terrace mapping. The comparison with the ground truth, as delineated by field survey, indicated that random forest performed best, with a 95.60% overall accuracy (followed by 94.16% and 92.33% for XGBoost and KNN, respectively). The influence of class imbalance and feature selection is discussed. This work provides a credible framework for mapping artificial terraces.

Quantifying the Influence of Achievement Emotions for Student Learning in MOOCs

2020 ◽  
pp. 073563312096731
Author(s):  
Bowen Liu ◽  
Wanli Xing ◽  
Yifang Zeng ◽  
Yonghe Wu
Keyword(s):  
Random Forest ◽  
Nearest Neighbor ◽  
Online Courses ◽  
Learning Performance ◽  
Support Vector ◽  
K Nearest Neighbor ◽  
Achievement Emotions ◽  
Integrative Framework ◽  
Emotional Interaction ◽  
Performance Results

Massive Open Online Courses (MOOCs) have become a popular tool for worldwide learners. However, a lack of emotional interaction and support is an important reason for learners to abandon their learning and eventually results in poor learning performance. This study applied an integrative framework of achievement emotions to uncover their holistic influence on students’ learning by analyzing more than 400,000 forum posts from 13 MOOCs. Six machine-learning models were first built to automatically identify achievement emotions, including K-Nearest Neighbor, Logistic Regression, Naïve Bayes, Decision Tree, Random Forest, and Support Vector Machines. Results showed that Random Forest performed the best with a kappa of 0.83 and an ROC_AUC of 0.97. Then, multilevel modeling with the “Stepwise Build-up” strategy was used to quantify the effect of achievement emotions on students’ academic performance. Results showed that different achievement emotions influenced students’ learning differently. These findings allow MOOC platforms and instructors to provide relevant emotional feedback to students automatically or manually, thereby improving their learning in MOOCs.

Symmetry Breaking and Training from Incomplete Data with Radial Basis Boltzmann Machines

1997 ◽  
Vol 08 (03) ◽  
pp. 301-315 ◽  
Author(s):  
Marcel J. Nijman ◽  
Hilbert J. Kappen
Keyword(s):  
Symmetry Breaking ◽  
Incomplete Data ◽  
Missing Values ◽  
Nearest Neighbor ◽  
Boltzmann Machine ◽  
K Nearest Neighbor ◽  
Data Set ◽  
Input Space ◽  
Learning Rules ◽  
Radial Basis

A Radial Basis Boltzmann Machine (RBBM) is a specialized Boltzmann Machine architecture that combines feed-forward mapping with probability estimation in the input space, and for which very efficient learning rules exist. The hidden representation of the network displays symmetry breaking as a function of the noise in the dynamics. Thus, generalization can be studied as a function of the noise in the neuron dynamics instead of as a function of the number of hidden units. We show that the RBBM can be seen as an elegant alternative of k-nearest neighbor, leading to comparable performance without the need to store all data. We show that the RBBM has good classification performance compared to the MLP. The main advantage of the RBBM is that simultaneously with the input-output mapping, a model of the input space is obtained which can be used for learning with missing values. We derive learning rules for the case of incomplete data, and show that they perform better on incomplete data than the traditional learning rules on a 'repaired' data set.

scholarly journals The K Nearest Neighbor Algorithm for Imputation of Missing Longitudinal Prenatal Alcohol Data

2021 ◽  
Author(s):  
Ayesha Sania ◽  
Nicolo Pini ◽  
Morgan Nelson ◽  
Michael Myers ◽  
Lauren Shuffrey ◽  
...  
Keyword(s):  
Missing Data ◽  
Missing Values ◽  
Nearest Neighbor ◽  
Learning Algorithm ◽  
Drinking Behavior ◽  
Nearest Neighbors ◽  
First Trimester ◽  
Epidemiologic Studies ◽  
K Nearest Neighbor ◽  
Timeline Followback

Abstract Background — Missing data are a source of bias in epidemiologic studies. This is problematic in alcohol research where data missingness is linked to drinking behavior. Methods — The Safe Passage study was a prospective investigation of prenatal drinking and fetal/infant outcomes (n=11,083). Daily alcohol consumption for last reported drinking day and 30 days prior was recorded using Timeline Followback method. Of 3.2 million person-days, data were missing for 0.36 million. We imputed missing data using a machine learning algorithm; “K Nearest Neighbor” (K-NN). K-NN imputes missing values for a participant using data of participants closest to it. Imputed values were weighted for the distances from nearest neighbors and matched for day of week. Validation was done on randomly deleted data for 5-15 consecutive days. Results — Data from 5 nearest neighbors and segments of 55 days provided imputed values with least imputation error. After deleting data segments from with no missing days first trimester, there was no difference between actual and predicted values for 64% of deleted segments. For 31% of the segments, imputed data were within +/-1 drink/day of the actual. Conclusions — K-NN can be used to impute missing data in longitudinal studies of alcohol use during pregnancy with high accuracy.

Evaluating Annotated Dataset of Customer Reviews for Aspect Based Sentiment Analysis

2021 ◽  
Author(s):  
Dimple Chehal ◽  
Parul Gupta ◽  
Payal Gulati
Keyword(s):  
Machine Learning ◽  
Support Vector Machine ◽  
Sentiment Analysis ◽  
Nearest Neighbor ◽  
Supervised Machine Learning ◽  
Support Vector ◽  
Product Reviews ◽  
K Nearest Neighbor ◽  
Customer Reviews ◽  
Percent Accuracy

Sentiment analysis of product reviews on e-commerce platforms aids in determining the preferences of customers. Aspect-based sentiment analysis (ABSA) assists in identifying the contributing aspects and their corresponding polarity, thereby allowing for a more detailed analysis of the customer’s inclination toward product aspects. This analysis helps in the transition from the traditional rating-based recommendation process to an improved aspect-based process. To automate ABSA, a labelled dataset is required to train a supervised machine learning model. As the availability of such dataset is limited due to the involvement of human efforts, an annotated dataset has been provided here for performing ABSA on customer reviews of mobile phones. The dataset comprising of product reviews of Apple-iPhone11 has been manually annotated with predefined aspect categories and aspect sentiments. The dataset’s accuracy has been validated using state-of-the-art machine learning techniques such as Naïve Bayes, Support Vector Machine, Logistic Regression, Random Forest, K-Nearest Neighbor and Multi Layer Perceptron, a sequential model built with Keras API. The MLP model built through Keras Sequential API for classifying review text into aspect categories produced the most accurate result with 67.45 percent accuracy. K- nearest neighbor performed the worst with only 49.92 percent accuracy. The Support Vector Machine had the highest accuracy for classifying review text into aspect sentiments with an accuracy of 79.46 percent. The model built with Keras API had the lowest 76.30 percent accuracy. The contribution is beneficial as a benchmark dataset for ABSA of mobile phone reviews.

scholarly journals Modeling Barrier Island Habitats Using Landscape Position Information

2019 ◽  
Vol 11 (8) ◽  
pp. 976
Author(s):  
Nicholas M. Enwright ◽  
Lei Wang ◽  
Hongqing Wang ◽  
Michael J. Osland ◽  
Laura C. Feher ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Nearest Neighbor ◽  
Learning Algorithms ◽  
Barrier Island ◽  
Barrier Islands ◽  
Machine Learning Algorithms ◽  
Landscape Position ◽  
K Nearest Neighbor ◽  
Island Habitats

Barrier islands are dynamic environments because of their position along the marine–estuarine interface. Geomorphology influences habitat distribution on barrier islands by regulating exposure to harsh abiotic conditions. Researchers have identified linkages between habitat and landscape position, such as elevation and distance from shore, yet these linkages have not been fully leveraged to develop predictive models. Our aim was to evaluate the performance of commonly used machine learning algorithms, including K-nearest neighbor, support vector machine, and random forest, for predicting barrier island habitats using landscape position for Dauphin Island, Alabama, USA. Landscape position predictors were extracted from topobathymetric data. Models were developed for three tidal zones: subtidal, intertidal, and supratidal/upland. We used a contemporary habitat map to identify landscape position linkages for habitats, such as beach, dune, woody vegetation, and marsh. Deterministic accuracy, fuzzy accuracy, and hindcasting were used for validation. The random forest algorithm performed best for intertidal and supratidal/upland habitats, while the K-nearest neighbor algorithm performed best for subtidal habitats. A posteriori application of expert rules based on theoretical understanding of barrier island habitats enhanced model results. For the contemporary model, deterministic overall accuracy was nearly 70%, and fuzzy overall accuracy was over 80%. For the hindcast model, deterministic overall accuracy was nearly 80%, and fuzzy overall accuracy was over 90%. We found machine learning algorithms were well-suited for predicting barrier island habitats using landscape position. Our model framework could be coupled with hydrodynamic geomorphologic models for forecasting habitats with accelerated sea-level rise, simulated storms, and restoration actions.

Sign in / Sign up

Export Citation Format

Share Document