Glaucoma which part of Diabetic Retinopathy is the disease which distorted the optical nerve system and resulted loss in vision. Fractal dimension is one of feature extraction that can been used in retinopathy fields due to the fractal dimension able to characterize the retinal vasculature. In this paper, we presented a research based on fractal dimension which not only distinguish the healthy subjects and diabetic retinopathy patients, but also severe level of diabetic retinopathy patients. By using MESSIDOR dataset and Random Forest as Classifier, we obtained the results that fractal dimensions are able to distinguish the healthy subjects and diabetic retinopathy patients, but it did not obtain satisfactory results for classifying the severity of diabetic retinopathy patients (grade level). Thus, future directions which need to be explore is the other features such as univariate, multivariate and other statistical features. It also needs to pay attention to red lesion detection to gain more information about diabetic retinopathy grade level.

Sentiment analysis in terms of polarity classification is very important in everyday life, with the existence of polarity, many people can find out whether the respected document has positive or negative sentiment so that it can help in choosing and making decisions. Sentiment analysis usually done manually. Therefore, an automatic sentiment analysis classification process is needed. However, it is rare to find studies that discuss extraction features and which learning models are suitable for unstructured sentiment analysis types with the Amazon food review case. This research explores some extraction features such as Word Bags, TF-IDF, Word2Vector, as well as a combination of TF-IDF and Word2Vector with several machine learning models such as Random Forest, SVM, KNN and Naïve Bayes to find out a combination of feature extraction and learning models that can help add variety to the analysis of polarity sentiments. By assisting with document preparation such as html tags and punctuation and special characters, using snowball stemming, TF-IDF results obtained with SVM are suitable for obtaining a polarity classification in unstructured sentiment analysis for the case of Amazon food review with a performance result of 87,3 percent.

Feature selection and ensemble learning can be used to improve the accuracy and robustness of epileptic seizure detection and classification. Unfortunately, few studies have fully utilized feature selection and ensemble learning. In this paper, we present an adaptive hybrid feature selection-based classifier ensemble (AHFSE) for epileptic seizure classification. The AHFSE creates a new sample subsets in every bootstrap using adaptive hybrid feature selection. It combines them using rank aggregation to obtain a distinguished subset of features. These new samples subsets are then fed into a classifier. Finally, majority voting is used to complete the detection and classification tasks. The AHFSE is designed to obtain an optimized subset of features based on the different samples in every bootstrap, which have a tendency to generate different results with respect to rank aggregation.With Discrete wavelet transform, the experiments based on binary and multi-class tasks show that the AHFSE performs well on the Bonn dataset and improves the specificity, sensitivity or accuracy of the selected features by combining the subsets of different feature selections to obtain new samples within the bagging process. Furthermore, the adaptive process helps the framework obtain the optimum combination of the feature selection algorithm. The AHFSE also obtains more desirable final results in several perspectives, such as: 1) compared with other feature selection methods; 2) compared with other ensemble methods and 3) compared with other research which using discrete wavelet transform as preprocessing step.

Data in the real world, there are many conditions (situations) where the number of instances in one class is much less than the number of instances in other classes. This situation is a problem in unbalanced datasets (imbalance class). As a result, performance in classification will decrease in some data systems. In this study, it was identified that the apple leaf disease performance dataset used had a large enough data imbalance problem where the comparison between instances was 1:5, so an oversampling method was needed to solve the data imbalance problem. Methods that can be used include the Synthetic Minority Over Sampling Technique (SMOTE). In order to validate the effectiveness of the proposed model, two experimental scenarios were carried out: first, the VGG16 algorithm was directly applied to modeling without considering class imbalance by reducing the number of layers and kernels in each layer to achieve optimal results, second, over-sampling SMOTE to increase the number of balanced datasets. The results showed that using the confusion matrix the accuracy results for each method were obtained where VGG 16 scored 85.16%, VGG 16 with SMOTE scored 92.94%. The conclusion of this study is that SMOTE helps improve the accuracy of leaf disease detection in apples.

Multi-label learning plays a critical role in the areas of data mining, multimedia, and machine learning. Although many multi-label approaches have been proposed, few of them have considered to de-emphasize the effect of noisy features in the learning process. To address this issue, this paper designs a new method named representative multi-label learning algorithm. Instead of considering all features, the proposed algorithm focuses only on the representative ones, via incorporating an affinity propagation algorithm, kernel formulation, and a multi-label support vector machine into the learning framework. Specifically, it first adopts an affinity propagation algorithm to select a set of representative features and capture the relationships among features. Then, the algorithm constructs the representative kernel functions to measure the similarity between data instances. Finally, a multi-label support vector machine is applied to solve the learning problem. Based on the representative multi-label learning algorithm, we further design a representative multi-label learning ensemble framework to improve the accuracy, stableness, and robustness. Experimental results show that the proposed algorithm works well on most of the datasets and outperforms the compared multi-label learning approaches.

Each company applies a contract extension to assess the performance of its employees. Employees with good performance in the company are entitled to future contracts within a certain period of time. In a pandemic time, many companies have made decisions to carry out WFH (Work from Home) activities even to Termination (Attrition) of Employment. The company's performance cannot be stable if in certain fields it does not meet the criteria required by the company. Thus, due to many things to consider in contract extension, we are proposed feature selection steps such as duplicate features, correlated features and Univariate Receiver Operating Characteristics curve (ROC) to reduce features from 35 to 21 Features. Then, after we obtained the best features, we applied into Decision Trees and Random Forest. By optimizing parameter selection using parameter grid, the research concluded that Random Forest with feature selection can predict Employee Attrition and Performance by obtain accuracy 79.16%, Recall 76% and Precision 82,6%.