Poverty is one of the problems prioritized for completion by the central government or the regions. This condition seems to have no limits because every year the problem of poverty is an issue that has always been a discussion of the government. As in Bali, even though the tourism industry is growing very rapidly, until now the problem of poverty is still a fundamental problem that needs to be resolved. Based on data from the Central Statistics Agency in 2016 the poverty rate of the province of Bali is 4.25% and one of the districts that has a higher poverty rate than the province is Tabanan Regency [1]. Various poverty alleviation programs have been implemented to break the cycle of poverty. However, poverty alleviation programs that have been implemented by the Tabanan regional government are still not optimal. In overcoming these problems, this study has the aim of creating an application system that can identify the conditions of households in Tabanan regency. The system built will identify a family based on 5 welfare categories so that it will provide an easy assessment for the poverty program survey officers. The system development model uses the K-Nearest Nighbor algorithm in modeling and classifying households. The results showed the system had an assessment accuracy rate of 83% Keywords: Poverty, Poor Households, K-Nearest Neighbor ABSTRAK Kemiskinan menjadi salah satu permsalahan yang diprioritaskan untuk di selesaikan oleh pemerintah pusat maupu daerah. Kondisi ini seakan tidak ada batasnya karena setiap tahun permasalahan kemiskinan merupakan isu yang selalu menjadi pembahasan pemerintah. Seperti halnya di provinsi bali, meskipun industri pariwisata berkembang sangat pesat namu sampai saat ini permasalahan kemiskinan masih menjadi permasalahan mendasar yang perlu diselesaikan. Berdasarkan data Badan Pusat Statistik tahun 2016 tingkat kemiskinan provinsi bali sebesar 4,25% dan salah satu kabupaten yang memiliki tingkat kemiskinan lebih tinggi dari provinsi adalah Kabupaten Tabanan [1]. Berbagai program pengentasan kemiskinan sudah dilaksanakan untuk memutus siklus kemiskinan yang terjadi. Namun program-program pengentasan kemiskinan yang sudah dilaksanakan pemerintah daerah Tabanan masih belum optimal. Dalam mengatasi permasalahan tersebut, pada penelitian ini memiliki tujuan untuk membuat sistem aplikasi yang dapat mengidentifikasi kondisi rumah tangga yang ada di kabupaten Tabanan. Sistem yang dibangun akan mengidentifikasi sebuah keluarga berdasarkan 5 katagori kesejahteraan sehingga akan memberikan kemudahan penilaian untuk petugas pendata program kemiskinan. Model pengembangan sistem menggunakan algoritma K-Nearest Nighbor dalam memodelkan dan mengklasifikasi rumah tangga. Hasil penelitian menunjukkan sistem memiliki tingkat akurasi penilaian sebesar 83% Kata Kunci : Kemiskinan, Rumah Tangga Miskin, K-Nearest Neighbor

Movies are an entertainment that is in great demand by many groups from children, teenagers, adults, and parents. In the current digital era, various films can be watched on television to digital streaming services. Public opinion on the films watched can be in the form of positive opinions or negative opinions. Sentiment analysis is one of the fields of Natural Language Processing (NLP) which is able to build a system to recognize and extract opinions in the form of text, sentiment analysis is usually used to find out people's opinions or assessments of a products, services, politics, or other topics. Through sentiment analysis from the collection of reviews, the public can get various recommendations for films that can be watched. The method implemented to classify review data into positive reviews and negative reviews in this study is LSTM by comparing two different optimizers, namely Adam and RMSprop. This study succeeded in providing sentiment predictions with different optimizers with accuracy values ??for the LSTM application with Adam Optimizer reaching 77.11% and the LSTM application with RMSprop reaching 80.07%.

Diagnosis adalah klasifikasi seseorang berdasarkan suatu penyakit atau abnormalitas yang diidapnya. Salah satu jenis penyakit yang memerlukan diagnosis adalah penyakit tumor otak. Akan tetapi dalam proses pemeriksaannya tentu memerlukan sautu biaya yang cukup mahal. Oleh sebab itu diperlukan suatu sistem yang dapat bertindak layaknya serang pakar untuk mengetahui gejala- gejala yang timbul akibat penyakit tumor otak. Salah satu metode yang dapat diterapkan dalam pembuatan sistem pakar adalah metode Certinty Factor(CF). Metode ini dapat digunakan untuk mengatasi permasalahan berkaitan dengan ketidakpastian dalam menyelesaikan atau menentukan suatu solusi. Hasil dari penelitian ini adalah prosantase kemungkinan pasien terhadap keempat jenis penyakit tumor otak yang ada.

Almost everyone looks at reviews before deciding to buy an item in e-commerce. Consumers say that online reviews influence their purchasing decisions. Based on these data, consumers need sentiment reviews to make a decision to choose a product/service. However, the results of the sentiment analysis are still less specific, so the review classification process is carried out based on the review category. Sentiment classification process based on clothing category is carried out using the Convolutional neural network method. The amount of data used is 3384 data with 3 categories. The category classification model shows good performance. When evaluated with testing data (unseen data), the accuracy value is 88%, the precision value is 88%, recall is 88% and the f1-score is 88%. For the sentiment classification model with the bottoms category, the resulting accuracy value is 80%, precision is 81%, recall is 80%, and f1-score is 79%. For the sentiment classification model with the dresses category, the accuracy value is 81%, precision is 81%, recall is 81%, and f1-score is 81%. For sentiment classification with the tops category the resulting accuracy value is 77%, precision is 77%, recall is 77%, and f1-score is 77%.

Classification is a process that automatically places text documents into a text based on the content of the text. Classification can help us classify many text documents that have been published, with the classification, these text documents can be reached easily and quickly. Feature selection can be used to improve the performance of text classification in terms of learning speed and effectiveness. In the Chi-Square feature selection experiment, a 1% threshold combination with a parameter value of k=6 is the combination chosen to be the best model. In testing the new data, the K-Nearest Neighbor model by selecting the Chi-Square feature produces precision performance, recall, F1-Score, and accuracy respectively, namely 85%, 83.3%, 88.2%, and 92.3%. In the Gini Index feature selection experiment,1% threshold combination with a parameter value of k=4 is the combination chosen to be the best model. This threshold selects about 31 features with the highest Gini Index value. In testing the new data, the K-Nearest Neighbor model by selecting the Gini Index feature produces precision performance, recall, F1-Score, and accuracy respectively, namely 81.2%, 80.3%, 81.6%, and 86.6%.