Rizal Setya Perdana
Fakultas Ilmu Komputer , Universitas Brawijaya

Published : 53 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Implementasi Fuzzy K-Nearest Neighbour (FK-NN) Untuk Pemilihan Keminatan Mahasiswa Teknik Informatika (Studi Kasus : Program Studi Teknik Informatika Fakultas Ilmu Komputer Universitas Brawijaya) Dhony Lastiko Widyastomo; Indriati Indriati; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 11 (2018): November 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1002.216 KB)

Abstract

Concentration selection is one of few steps for a students to finish their studies. Informatics programee have 4 concentration consist of Artificial Inteligent(AI), Software Engineering (SE), Network and Game. Unfortunately because the limited and many internal problems from the students causing some problem for the concetration selection. To solve the problem of selection, a system who can give a classification is needed to give the solution. A classification for concentracion selection uses fuzzy k-nearest neighbor for its method. The method works with calculate the number of K value to Process the classification of 4 study concentration and resulting the recomendation class of concentration class based on the student data. Based on the research of study using 200 data of the students of Informatics engineering, from 2011 to 2013, the biggest accuracy was produced by K value=3 and have 87,5% accuracy. While the lowest precentages of accuracy was produced by K value=10 with the averages of 67,5% accuracy.
Analisis Sentimen Cyberbullying pada Komentar Instagram dengan Metode Klasifikasi Support Vector Machine Wanda Athira Luqyana; Imam Cholissodin; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 11 (2018): November 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1230.493 KB)

Abstract

Instagram is the most popular social media in these recent days. The users who start from kids, teenagers to adults, have the role in boosting the popularity of Instagram. However, this social media could not be seperated from the dangers of cyberbullying which is done often by the users, especially in the comment column. The dangers of cyberbullying are certainly worried many people because of the impact it has. Therefore, a sentiment analysis in Instagram comment column can be done in order to find out the sentiments in each comment. Sentiment analysis is a branch of text mining science which is used to extract, understand, and cultivate the data. This research used Term Frequency-Inverse Document Frequency (TF-IDF) and Support Vector Machine (SVM) classification method to examine the sentiments in each comment. Data consisted of 400 data which taken offline have a total 1799 features. The comment document is divided into 70% of training data and 30% of test data. Based on the tests performed, the best parameters obtained in the SVM method are the degree of polynomial kernel 2, the average of learning rate of 0.0001, and the maximum number of iterations which is 200 times. From these result, it obtained that the highest accuracy is 90%, 50% in the training data composition and 50% composition of test data.
Peringkasan Teks Otomatis Menggunakan Metode Maximum Marginal Relevance Pada Hasil Pencarian Sistem Temu Kembali Informasi Untuk Artikel Berbahasa Indonesia Nirmala Fa'izah Saraswati; Indriati Indriati; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 11 (2018): November 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1023.851 KB)

Abstract

Information retrieval is a system that displays documents according to the query given by user. However, the information retrieval system provide a lot of search results, when we are looking for a desired information is not possible to open one by one documents generated by system. Text Summarization can be done to get an overview of information from a document, so that user get the right documents. One method to summarize text is Maximum Marginal Relevance (MMR). Maximum Marginal Relevance (MMR) is one of the extractive summary methods used to summarize single or multi document documents. MMR summarizes documents by computing the similarity between sentences and sentences, and between sentences and queries. Based on the test results, it obtain best Precision at k in the fifth rank of 0.96 for information retrieval system results. The best test results from an average precision, recall, f-measure and accuracy respectively 0.70, 0.75, 0.70 and 74.17. The used method is good enough to get the relevant documents and obtain summaries based on the title corresponding to the contents of the document.
Prediksi Indeks Harga Konsumen (IHK) Kelompok Perumahan, Air, Listrik, Gas, dan Bahan Bakar Menggunakan Metode Extreme Learning Machine Irma Ramadanti Fitriyani; Budi Darma Setiawan; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 11 (2018): November 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1131.86 KB)

Abstract

The Consumer Price Index is one of the indicators to measure the inflation rate in Indonesia. In 2017 inflation in Indonesia by expenditure group in general is 3,61%. The group of housing, water, electricity, gas, and fuel become the biggest contributor of inflation compared to six other groups with 5,14%. Therefore, the prediction needs to be done to anticipate and reduce domestic inflation rate. Prediction done in this research using method of Extreme Learning Machine (ELM) with initialization of weight using Nguyen-Widrow initialization. The data used in this research are 84 Consumer Price Index data of housing, water, electricity, gas, and fuel from January 2011 until December 2017. The data obtained from the official website of Bank Indonesia (www.bi.go.id). The result of this research is the minimum RMSE value of 0,72 with the number of features = 7, the amount of training data is 30 and the testing data is 11, the number of hidden neurons = 7, and the activation function is sigmoid binary function.
Penentuan Kenaikan Jabatan Karyawan Menggunakan Metode Fuzzy-Analytical Hierarchy Process (FAHP) di Pabrik Gula Lestari Patianrowo Nganjuk Erma Rafliza; Indriati Indriati; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 12 (2018): Desember 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1004.207 KB)

Abstract

Career promotion is an important factor and heavily expected by the employees. Establishment of position that analyzed manually will be very disadvantage if there is Human Error or error conducted by humans related to the careful analysis towards the process compared to calculation conducted by machine. The implementation of Fuzzy-Analytical Hierarchy Process (FAHP) is conducted by determining matrix values of paired comparison as parameter about how important the parameter compared to other parameters. Three parameters as comparison are KIN (Individual Competence), KT (Core Competence), and KP (Role Competence). In this research, it used total test data of 78 which obtained result with accuracy level in the created system is 92.3%, in which matrix values of paired comparison as with expert statement, then if the level of importance for a parameter replaced to be slightly more important than value established by expert then accuracy level will be better, which is 94.87%. However, if the values established by expert replaced to be the opposite values then the obtained accuracy level is very low, which is only 79.48%. Therefore, although accuracy level by expert is already good, however, if a parameter changed to be slightly more important, then, the obtained result will be more optimal.
Implementasi Algoritme Modified K-Nearest Neighbor (MKNN) untuk Diagnosis Penyakit Tanaman Cengkeh Rizaldy Amsyar; Nurul Hidayat; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 12 (2018): Desember 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (400.455 KB)

Abstract

Clove plant have high economic value and one of many export commodity of Indonesian plantation product, in Wonosalam region Jombang Regency there are less well groomed clove farm because the owners are not at all the times in the farm, and thus the plant susceptible to disease and reduced yields from the clove harvest. Needed a way to help farmers to know the types of diseases that attack the clove plants, then made a clove plant diagnosis system using the algorithm Modified K - Nearest Neighbor (MKNN). The diagnostic system will provide clove plant disease information based on inputs of observable symptoms of the plant. MKNN algorithm is the development of KNN algorithm by adding calculation process of data training validation and weight voting. Validation calculation aims to overcome the problem of data that deviates on the KNN algorithm in order to avoid bias and weight voting aims to calculate the weight of the data. Accuracy of clove plant diagnosis system using MKNN algorithm is 96.67%.
Klasifikasi Berita Pada Twitter dengan Menggunakan Metode Naive Bayes dan Feature Expansion Berbasis Cosine Similarity Resti Febriana; Mochammad Ali Fauzi; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 12 (2018): Desember 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (399.899 KB)

Abstract

Information has become indispensable in this modern era, especially with the existence of various social media that support information update. Twitter as one of the most active social media is used to update information belonging to short text or short stories that have some difficulty when done classification, such as ambiguous word, the word contained in the test data never appear in the data train and so on. This research was conducted to determine the effect of using feature expansion or addition of word on short text in the result of classification. Prior to classification, the first data to be tested is added to the list of pre-made words as an external source or dictionary with specified limits. This limitation aims to determine the minimum value of the most optimal limit in generating the highest accuracy in the classification process. In the process of making external sources cosine similarity process is done to find the closeness between words. The result of this research is accurate showing effect of expansion of feature expansion in classification result, 83% accuracy in classification without feature expansion and increased to 87% on feature expansion with threshold value 0.9.
Klasifikasi Video Clickbait pada YouTube Berdasarkan Analisis Sentimen Komentar Menggunakan Learning Vector Quantization (LVQ) dan Lexicon-Based Features Dwi Wahyu Puji Lestari; Rizal Setya Perdana; Putra Pandu Adikara
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 2 (2019): Februari 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (967.888 KB)

Abstract

Clickbait is social media content that aims to attract website visitors in order to visit their content by creating clickbait in form of appealing or provoking title but with irrelevant content. It makes the visitor decieved and disappointed, so they usually vent their frustation by writing their positive or negative opinion on the comment section. The document that is used in the research comes from YouTube comments that is related with Indonesian clickbait and non-clickbait content. This research used Learning Vector Quantization (LVQ) method and Lexicon-Based Features as word weighting other than using TF-IDF. This research uses 300 data consisting 2 type of data, training and testing data with the ratio of 70% training data and 30% testing data. The accuracy of the system that is obtained by classification using LVQ without Lexicon-Based Features is 54.54%, 1 precission, 0.1667 recall and 0.2858 f-measure. The result of the accuracy of the system using LVQ and Lexicon-Based Features is 90.91%, 0.8571 precission, 1 recall, and 0.9231 f-measure. The conclution is that LVQ method and Lexicon-Based Features can be used for sentiment classification.
Klasifikasi Kemacetan Lalu Lintas di Kota Malang Pada Sosial Media Twitter Menggunakan Metode Improved K-Nearest Neighbor Riska Dewi Nurfarida; Indriati Indriati; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 2 (2019): Februari 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (809.932 KB)

Abstract

Twitter is a social media network that has many users that can be used for communication media. And from Twitter you can also get various forms of information including negative and positive opinions and various other types of information. One of the information that can be obtained from Twitter is information about traffic conditions. Malang City community uses Twitter social media as one of the media to get information about traffic conditions. Through the @PuspitaFM account, the people of Malang City share information about the state of traffic around them. From the @PuspitaFM account, every day I will share tweets about traffic conditions in Malang City either by tweeting directly or tweets from followers that will be retweeted by the @PuspitaFM account. Of all the tweets that exist, sometimes there is confusion that occurs in the categorization of traffic jams or not jammed in the tweet. Therefore, the classification of tweets is jammed or not jammed as a solution to the problem. There are several processes carried out in this study, namely starting from prepocessing text which is divided into cleansing, case folding, tokenisation, filtering and stemming processes. The process will continue with the term weighting or weighting process, followed by normalization, cosine similiarity and classification processes with the Improved K-NN method. The results obtained from this study are recall value of 0.42857, precision value of 0.71428, f-measure value of 0.53571 and the best accuracy of 65.33%. The training data used is 600 tweet documents, and 150 test data tweet documents.
Klasifikasi Kemacetan Lalu Lintas Kota Malang Melalui Media Twitter Menggunakan Metode Neighbor Weighted K-Nearest Neighbor (NW-KNN) Putu Amelia Vennanda Widyaswari; Indriati Indriati; Rizal Setya Perdana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 2 (2019): Februari 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (481.22 KB)

Abstract

Twitter is a social media that is still widely used today. Like other social media Twitter is useful for making friends, exxchanging messages, and information about various things such as entertainment, economy, politics, and so forth. Twitter is also useful for finding information about the state of traffic on a road by accessing traffic accounts on Twitter. However, tweets are often found with ambigous words about the condition of the road. So tweets needs to classified to make it easy for road users. Classification begins with doing preprocessing stages on training and test documents, then proceeding with weighting TF-IDF until the classification stage using the NW-KNN (Neighbor Weighted K-Nearest Neighbor) method. Based on the implementation and testing carried out on the study of Malang City Traffic Congestion Classification Through Media Twitter Using Neighbor Weighted K-Nearest Neighbor (NW-KNN) method which uses 600 training data and 150 test data, obtained results of 0.7336507 for the average precision, 0.2210526 for recall, 0.3002686 for f-measure, and accuracy obtained at 0.665.