To avoid citizen disputes, hate speech on social media, such as Twitter, must be automatically detected. The current research in Indonesian Twitter focuses on developing better hate speech detection models. However, there is limited study on the explainability aspects of hate speech detection. The research aims to explain issues that previous researchers have not detailed and attempt to answer the shortcomings of previous researchers. There are 13,169 tweets in the dataset with labels like “hate speech” and “abusive language”. The dataset also provides binary labels on whether hate speech is directed to individual, group, religion, race, physical disability, and gender. In the research, classification is performed by using traditional machine learning models, and the predictions are evaluated using an Explainable AI model, such as Local Interpretable Model-Agnostic Explanations (LIME), to allow users to comprehend why a tweet is regarded as a hateful message. Moreover, models that perform well in classification perceive incorrect words as contributing to hate speech. As a result, such models are unsuitable for deployment in the real world. In the investigation, the combination of XGBoost and logical LIME explanations produces the most logical results. The use of the Explainable AI model highlights the importance of choosing the ideal model while maintaining users’ trust in the deployed model.

Diabetic retinopathy is the leading cause of blindness suffered by working-age adults. The increase in the population diagnosed with DR can be prevented by screening and early treatment of eye damage. This screening process can be conducted by utilizing deep learning techniques. In this study, the detection of DR severity was carried out using the hybrid CNN-DELM method (CDELM). The CNN architectures used were ResNet-18, ResNet-50, ResNet-101, GoogleNet, and DenseNet. The learning outcome features were further classified using the DELM algorithm. The comparison of CNN architecture aimed to find the best CNN architecture for fundus image features extraction. This research also compared the effect of using the kernel function on the performance of DELM in fundus image classification. All experiments using CDELM showed maximum results, with an accuracy of 100% in the DRIVE data and the two-class MESSIDOR data. Meanwhile, the best results obtained in the MESSIDOR 4 class data reached 98.20%. The advantage of the DELM method compared to the conventional CNN method is that the training time duration is much shorter. CNN takes an average of 30 min for training, while the CDELM method takes only an average of 2.5 min. Based on the value of accuracy and duration of training time, the CDELM method had better performance than the conventional CNN method.

The linear algebra applications available today usually only provide the result. So, it is a challenge to overcome this, and innovation is needed in the computing aspect. One of the popular and open-source programming languages ​​is R. The computational innovation in R needs to be explored further, to explore the R programming logic. The creation of a function environment with the list function and the involvement of local and global variables/objects has received little attention. Based on the problems formulated, this study proposes two objectives, namely (1) developing an R program that is able to provide interactive and step-by-step solutions, to obtain a solution of a system of linear equations, and (2) to explore R’s ability to create and handle global variables. An R program is created, starting with creating a function environment. This function environment is filled with four related functions, namely “exchange”, “multiply”, “fold”, and “yield”. These four functions are connected to each other through a global object. Users can type in each function to perform row/column operations, interactively and step by step. The environmental function in this program, is named OBE. The OBE function accepts input in the form of a coupling matrix derived from a system of linear equations. The final result of this interactive process chain is given by the “result” function. The result function will display two matrices, namely the Original Matrix which is the input and the Equivalent Matrix.

Stunting is regarded as a harmful and concerning health issue in many developing countries. According to the World Health Organization, Indonesia ranks third in Southeast Asia in terms of stunting prevalence, with East Nusa Tenggara having the highest prevalence rate, 40.3%. The number of children is increasing every year and it is related to nutritional status because stunting’s number of children is quite high. To reduce that case, we do the research for analysis of factors affecting several child stunting and developing web applications. The data that we used for analysis is from Rote Ndao Regency, East Nusa Tenggara. Methods of the analysis are using Zero-Inflated Poisson Regression to build a model, analysis of factor that affected child stunting. A full model is built from six variables and three of them are significant from log model and two from logit model. However, the model is the best model with the smallest AIC 484.5389 and included four variables in the final model.

Insurance companies have business risks, one of which is credit risk. This study predicts the credit risk of prospective customers using Bayesian survival analysis since we used various type of variables in this paper, consist of nine categorical variables and one numerical variable. The modeling is carried out using insurance company data and through several experiments with the R software. Due to the use of the estimation method in this package using Reversible Jump Markov Chain Monte Carlo (RJMCMC), the resulting parameter coefficient (beta) will be different each time the simulation is carried out. To get a better beta value, several runs were carried out with different number of iterations and burn-up positions. The number of iterations and burn-up positions tried were as follows (60000,1000), (60000, 59000), (100000, 1000), (100000, 99000), (10000, 1000), (10000, 1000), (10000, 100), (200000, 100). Based on the output beta value from the run, it was found that the beta value was quite stable when the number of iterations was 100000 with a burn-up position of 1000. Factors that influence the occurrence of policy lapse are gender, place of residence, amount of insurance premium, period of premium payment, method of payment of premiums, and investment instruments.