Freedom of opinion through social media is frequently affect a negative impact that spreads hatred. This study aims to automatically detect Indonesian tweets that contain hate speech on Twitter social media. The data used amounted to 4,002 tweets related to politics, religion, ethnicity and race in Indonesia. The application model uses classification methods with machine learning algorithms such as Naïve Bayes, Multi Level Perceptron, AdaBoost Classifier, Decision Tree and Support Vector Machine. The study also compared the performance of the model using SMOTE to overcome imbalanced data. The results show that the Multinomial Naive Bayes algorithm produces the best model with the highest recall value of 93.2% which has an accuracy value of 71.2% for the classification of hate speech. Therefore, the Multinomial Naïve Bayes algorithm without SMOTE is recommended as the model to detect hate speech on social media.

Academic loads in writing a thesis cause difficulties to students to complete their studies. This problem can be solved with the application of an expert system with early detection of depressive final-year students that compares Certainty Factor method and Dempster-Shafer method. Certainty Factor method used to measure facts certainty or rules in defining the level of expert confidence, while the Dempster-Shafer method combines uncertainty that has characteristics according to the way of thinking of experts but has a mathematical basis. The purpose of comparing these methods is to get the most appropriate and best method for early detection of depression. The methodology used in this study is a comparative test between two methods, with a mechanism (1) Collecting data related to the depression of final year students; (2) Data acquisition and weighting of confidence values; (3) Calculate formulations from both methods; (4) Perform a comparative test to obtain conclusions. Certainty Factor method is the most appropriate and the best model in early detection of depression, because Certainty Factor method revealed that the more symptoms are given, the higher the probability of detection of depression.

Data confidentiality is an important aspect of information systems. In case of that we need an application to maintain data confidentiality. This study aims to compare two methods of data security systems, namely using the RC4A algorithm which is a symmetric key cryptographic algorithm and the RSA algorithm which is an asymmetric algorithm with a hybrid process on the RC4A and RSA algorithms to secure the secret key of RC4A and speed up the encryption process of RSA. The study uses a comparative test on two algorithm methods, according to data security system parameters. Data is encrypted with the RC4A algorithm and the RC4A key obtained from KSA RC4A will be encrypted with RSA before sending to the receiver. The test results get Hybrid cryptosystem RC4A-RSA faster in completing the encryption process than RSA. Furthermore, the RC4A key can be secured by being encrypted using the RSA algorithm.

Background: Stunting problems in toddlers, which until now are still a special concern in the world because it can inhibit the physical and mental development of children. Stunting is a chronic malnutrition problem caused by a lack of nutrient intake over a long period of time due to food intake that is not in accordance with the nutritional needs of the body. Toddlers who experience stunting will cause increased risk and stunted growth and development of children. The method used in this study was observational analytic with a controud case design. Sampling using proportional random sampling technique, obtained 86 respondents in accordance with the inclusion criteria, data collection was done using questionnaire sheets and using KMS to observe birth weight in infants, height and weight of the last weighing at the time of the study with questionnaire filling. The collected data is then processed by using the Chi Square test with α <0.05. The results of the statistical test showed that BBLR  (p = 0.005), Exclusive breastfeeding (p = 0.104), Mp-ASI (p = 0.121). The conclusion in this study is that BBLR affects the incidence of stunting, Exclusive breastfeeding and breastfeeding does not affect the incidence of stunting in children aged 12 to 59 months at the Campalagian Health Center in Polewali Mandar district.

Abstrak: Perkembangan ilmu pengetahuan dan teknologi yang semakin pesat mengharuskan banyak sektor dalam pendidikan melakukan perubahan. Pembelajaran blended learning adalah contoh dari salah satu pemanfaatan perkembangan teknologi informasi yang dimanfaatkan dalam sektor Pendidikan. Kombinasi pembelajaran dengan sistem tatap muka dan pembelajaran berbasis komputer yang dilakukan secara daring (dalam jaringan) maupun luring (luar jaringan) adalah bentuk pembelajaran blended learning Flex termasuk dalam jenis pembelajaran metode Blended Learning di mana proses belajar mengajar dilakukan dalam bentuk daring namun masih melakukan pembelajaran luring sebagai pendukung. Tujuan dari penelitian ini untuk melihat dampak dan manfaat dari penerapan flex blended learning dalam proses pembelajaran sehingga dapat mempermudah proses pelaksanaan pendidikan bagi guru dan juga siswa agar tujuan pendidikan dapat dicapai lebih maksimal. Data hasil penelitian diolah dengan menghitung nilai n-gain dan uji t untuk mengetahui dampak dari pembelajaran menggunakan flex blended learning. Hasil menunjukkan nilai n-gain siswa yang melakukan pembelajaran flex blended learning adalah 0,74 lebih tinggi daripada pembelajaran konvensional yang memiliki nilai 0,54. Uji t menunjukkan nilai sig. 0,00000018 dan nilai thitung 6,78 yang berarti pembelajaran menggunakan flex blended learning menunjukkan perbedaan yang signifikan terhadap pembelajaran konvensional. Dari hasil nilai n-gain dan uji t dapat ditarik kesimpulan bahwa pembelajaran flex blended learning dapat meningkatkan nilai hasil belajar siswa.Kata Kunci: Pembelajaran, Blended Learning, Flex Blended Learning.Abstract: The rapid development of science and technology requires many sectors in education to make changes. Blended learning is an example of one of the uses of information technology developments that are used in the education sector. The combination of learning with face-to-face systems and computer-based learning conducted online (in the network) and offline (outside the network) is a form of blended learning. Flex is included in the type of Blended Learning model where learning is carried out online but still does offline learning as a support. The purpose of this study is to see the impact and benefits of implementing flex blended learning in the learning process so that it can facilitate the process of implementing education for teachers and students so that educational goals can be achieved more optimally. The research data were processed by calculating the n-gain value and t-test to determine the impact of learning using flex blended learning. The results show that the n-gain value of students who do flex blended learning is 0.74 higher than conventional learning which has a value of 0.54. The t test shows the value of sig. 0.00000018 and t-count 6.78 which means learning using flex blended learning shows a significant difference to conventional learning. From the results of the n-gain value and t test, it can be concluded that flex blended learning can increase the result of student learning outcomes..Keywords: Learning, Blended Learning, Flex Blended Learning.