Claim Missing Document
Check
Articles

Found 10 Documents
Search

SENTIMENT ANALYSIS ON TWITTER BY USING MAXIMUM ENTROPY AND SUPPORT VECTOR MACHINE METHOD Cindo, Mona; Rini, Dian Palupi; Ermatita, Ermatita
SINERGI Vol 24, No 2 (2020)
Publisher : Universitas Mercu Buana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22441/sinergi.2020.2.002

Abstract

With the advancement of social media and its growth, there is a lot of data that can be presented for research in social mining. Twitter is a microblogging that can be used. In this event, a lot of companies used the data on Twitter to analyze the satisfaction of their customer about product quality. On the other hand, a lot of users use social media to express their daily emotions. The case can be developed into a research study that can be used both to improve product quality, as well as to analyze the opinion on certain events. The research is often called sentiment analysis or opinion mining. While The previous research does a particularly useful feature for sentiment analysis, but it is still a lack of performance. Furthermore, they used Support Vector Machine as a classification method. On the other hand, most researchers found another classification method, which is considered more efficient such as Maximum Entropy. So, this research used two types of a dataset, the general opinion data, and the airline's opinion data. For feature extraction, we employ four feature extraction, such as pragmatic, lexical-grams, pos-grams, and sentiment lexical. For the classification, we use both of Support Vector Machine and Maximum Entropy to find the best result. In the end, the best result is performed by Maximum Entropy with 85,8% accuracy on general opinion data, and 92,6% accuracy on airlines opinion data.
Efficient mobilenet architecture as image recognition on mobile and embedded devices Khasoggi, Barlian; Ermatita, Ermatita; Samsuryadi, Samsuryadi
Indonesian Journal of Electrical Engineering and Computer Science Vol 16, No 1: October 2019
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v16.i1.pp389-394

Abstract

The introduction of a modern image recognition that has millions of parameters and requires a lot of training data as well as high computing power that is hungry for energy consumption so it becomes inefficient in everyday use. Machine Learning has changed the computing paradigm, from complex calculations that require high computational power to environmentally friendly technologies that can efficiently meet daily needs. To get the best training model, many studies use large numbers of datasets. However, the complexity of large datasets requires large devices and requires high computing power. Therefore large computational resources do not have high flexibility towards the tendency of human interaction which prioritizes the efficiency and effectiveness of computer vision. This study uses the Convolutional Neural Networks (CNN) method with MobileNet architecture for image recognition on mobile devices and embedded devices with limited resources with ARM-based CPUs and works with a moderate amount of training data (thousands of labeled images). As a result, the MobileNet v1 architecture on the ms8pro device can classify the caltech101 dataset with an accuracy rate 92.4% and 2.1 Watt power draw. With the level of accuracy and efficiency of the resources used, it is expected that MobileNet's architecture can change the machine learning paradigm so that it has a high degree of flexibility towards the tendency of human interaction that prioritizes the efficiency and effectiveness of computer vision.
ASSOCIATION RULE METHOD FOR INFORMATION SYSTEM EPIDEMIC DENGUE MAPPING BASED ASSOCIATION OF RISK FACTORS IN PALEMBANG Ermatita, Ermatita; Destriatania, Suci
Prosiding International conference on Information Technology and Business (ICITB) 2015: INTERNATIONAL CONFERENCE ON INFORMATION TECHNOLOGY AND BUSINESS (ICITB) 1
Publisher : Prosiding International conference on Information Technology and Business (ICITB)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Endemic diseases dangerous such as dengue fever must be handling seriously  for the risk minimize by the disease. Dengue Hemorrhagic Fever (DHF) is  disease  has not been found vaccine or cure is powerful. It is  necessary  treatment to prevent the occurrence of dengue fever, especially when it came to the incidence of dengue fever endemic in certain areas by doing Epidemiologist dengue fever. Epidemiology is identification of risk factors for DHF to find level of area  risk. Risk factors of hemorrhagic fever endemic must be identified  to prevent the occurrence of dengue fever. Identifying risk factors and  risk factors association  can  potential increase  the occurrence of dengue fever. This study developed  mapping information system Dengue epidemic through Association rule method of data mining. The information generated in the map of epidemic DHF level based association of potential risk factors that cause hemorrhagic fever endemic. Analysis with the Association Rule to determine level of  DHF epidemic area based data reporting system. KEY WORDS: information system mapping, data mining, Association Rule, endemic, Dengue Hemorrhagic Fever (DHF).
IMPLEMENTASI SISTEM PAKAR PADA PASIEN PENDERITA TUBERKULOSIS POTENTIAL DROP OUT DI RUMAH SAKIT CUT MEUTIA ACEH UTARA Darnila, Eva; Ula, Mutammimul; Mauliza, Mauliza; Ermatita, Ermatita; Pahendra, Iwan
KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Vol 2, No 1 (2018): Peranan Teknologi dan Informasi Terhadap Peningkatan Sumber Daya Manusia di Era
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (23.348 KB) | DOI: 10.30865/komik.v2i1.968

Abstract

The existence of a technology that identifies and controls patients with potential drop out TB disease which is increasingly rapid will be a top priority, especially for the health team in following up the success of treatment. In this study, an expert system was used to diagnose patients with potential Drop Out tuberculosis by using a Case Based Reasoning model to see patients with potential Droup Out. For variable names used are pulmonary smear patients (+), new patients, pulmonary smear (-) / ro (+), new patients, extra pulmonary, relapsed patients, re-treatment, default patients, re-treatment patients, failed patients and others -other. The last detection process is taken from the highest value obtained in the diagnosis of all the symptoms that have been witnessed. Based on the results of the application of the Expert System on Potential Drop Out Tuberculosis Patients at Cut Meutia Hospital in North Aceh based on the case code 31 with a detection system for the AFB (+) Lung Patient with its detection symptoms, the patient coughs with phlegm for 2-3 weeks or more. the results of sputum examination, patients who have been treated with TB drugs less than 1 month and TB patients on sputum examination, patients who have been treated with TB drugs less than 1 month, TB patients stop the treatment and TB patients return to the facility health service facilities with the highest case value of 0.6111 of all detection systems that have been tested.Keywords: Expert system,  CBS, TB
ANALISIS MODEL NAIVE BAYES UNTUK IDENTIFIKASI PENGGOLONGAN DAYA LISTRIK DI KOTA LHOKSUMAWE Sadli, Muhammad; Fajriana, Fajriana; Fuadi, Wahyu; Ermatita, Ermatita; Pahendra, Iwan
KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Vol 2, No 1 (2018): Peranan Teknologi dan Informasi Terhadap Peningkatan Sumber Daya Manusia di Era
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (23.348 KB) | DOI: 10.30865/komik.v2i1.971

Abstract

Electricity subsidy is provided for all 450 VA power household customers and 900 VA power household customers who are poor and disadvantaged. However, there are many facts that household customers with 450 VA power are capable and 900 VA power household customers consist of capable households, boarding houses or luxury rented. Households are able to use more electricity than poor households. This paper describe to the identification of household customers' electrical power in the Lhokseumawe city to facilitate PLN in classifying customer power by using the Naive Bayes method. Naive bayes value variables used in this study are: monthly income, highest diploma, last job, house area, subscription fee and government registered household. The classification of household customer power is grouped into three categories, namely low (450 VA down), medium (900 VA) and high (above 1300 VA).. Based on household customer data that is used as training data, the Naive Bayes method is able to classify the customer data tested. So the Naive Bayes method successfully predicts the magnitude of the probability of household electrical power with an accuracy percentage of 80%.Keywords: Electricity, Naive Bayes,  CBS, low birth weight, subsidy
PERAMALAN POTENSI KEBAKARAN HUTAN DENGAN METODE DATA MINING Ermatita, Ermatita; Sukemi, Sukemi; Pratomo, Yudha
SEINASI-KESI Vol 1, No 1 (2018): Seinasi-Kesi 2018
Publisher : Fakultas Ilmu Komputer UPN Veteran Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (173.163 KB)

Abstract

Kebakaran hutan merupakan kejadian yang selalu berualang setiap tahun, terutama di Sumatera Selatan. Penanggulangan kebakaran hutan ini, terutama di lahan gambut, perlu dianalisis agar dapat diprediksi pola penyebaran kebakaran lahan gambut. Hal ini perlu dilakukan karena pemadaman kebakaran lahan gambut sulit dilakukan.Penelitian ini memproses data faktor-faktor yang memberikan potensi kebakaran hutan menerapkan data mining. Metode yang digunakan dalam penelitian ini . Hasil dari penelitian ini menunjukkan bahwa faktor-faktor penyebab dari kebakaran hutan yang diproses mempunyai potensi yang sama dalam menyebabkan kebakaran hutan.
Penentuan Prioritas Pengembangan Industri Kecil dan Menengah di kota Palembang Metode Weighted Product (WP) (studi kasus : Dinas Perindustrian, Perdagangan dan Koperasi Kota Palembang Ermatita, Ermatita; Zalika, Indah; Putra, Pacu
SEINASI-KESI Vol 2, No 1 (2019): Seinasi-Kesi 2019
Publisher : Fakultas Ilmu Komputer UPN Veteran Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (919.074 KB)

Abstract

Dinas Perindustrian, Perdagangan dan Koperasi (Disperindag) Kota Palembang merupakan salah satu unsur pemerintahan yang bertanggung jawab di bidang pengembangan Industri Kecil Menengah(IKM) Kota Palembang. Dalam pengembang IKM sendiri terkadang sulit menentukan IKM yang layak disebut prioritas karena banyaknya kriteria yang sama untuk penilaian pada IKM Kota Palembang serta belum adanya sistem yang memiliki pemodelan data untuk mendukung keputusan tersebut. Untuk meminimalkan kendala tersebut maka diperlukan suatu sistem pendukung keputusan yang dapat membantu para pengambil keputusan(decision maker) menganilisa IKM yang layak mendapat prioritas. Salah satu metode yang dapat digunakan yaitu weighted product (WP), metode weighted product dapat digunakan untuk proses keputusan multi-dimensi, yang disebut dengan analisis berdimensi. Dalam pengembangan sistem ini  menggunakan metode waterfall, waterfall dipilih karena stakeholder mendefinisikan terlebih dahulu nilai bobot yang diperlukan dalam proses perhitungan metode WP.  Dari hasil implementasi sistem, disimpulkan bahwa dengan penggunaan sistem ini dapat membantu proses pengambilan keputusan untuk proses penentuan prioritas pengembangan IKM di Kota Palembang
Model Promosi Pemilihan Jabatan Manajer Menggunakan Metode Weighted Product (WP) ( Studi Kasus: Bank Indonesia Provinsi Sumatera Selatan) Ermatita, Ermatita; Safithri, Selviana Rizki; Bardadi, Ali; Adrezo, Muhammad
SEINASI-KESI Vol 3, No 1 (2020): SEINASI-KESI 2020
Publisher : Fakultas Ilmu Komputer UPN Veteran Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Promosi jabatan memiliki peranan penting sebagai salah satu pergerakan pertumbuhan perusahaan. Bank indonesia provinsi sumatera selatan merupakan salah satu tempat mempunyai fungsi penting di dalam melayani masyarakat terutama masyarakat yang berada di provinsi sumatera selatan. Dalam pemilihan jabatan manajer terkadang sulit untuk menentukan siapa yangg layak dijadikan manajer karena banyaknya kriteria yang sama untuk penilaian jabatan manajer serta belum adanya sisstem yang memiliki permodelan data untuk mendukung suatu sistem pendukung keputusan. Untuk meminimalkan kendala tersebut maka diperlukan suatu sistem pendukung keputusan yang dapat membantu para pengambil keputusan (decision maker) menganalisa promosi jabatan manajer yang layakmendapatkan prioritasnya. Salah satu metode yang digunakan yaitu metode weighted product (WP) dimana metode weighted product adalah metode pengambilan keputusan multi kriteria
IMPLEMENTASI SISTEM PAKAR PADA PASIEN PENDERITA TUBERKULOSIS POTENTIAL DROP OUT DI RUMAH SAKIT CUT MEUTIA ACEH UTARA Darnila, Eva; Ula, Mutammimul; Mauliza, Mauliza; Ermatita, Ermatita; Pahendra, Iwan
KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Vol 2, No 1 (2018): Peranan Teknologi dan Informasi Terhadap Peningkatan Sumber Daya Manusia di Era
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/komik.v2i1.968

Abstract

The existence of a technology that identifies and controls patients with potential drop out TB disease which is increasingly rapid will be a top priority, especially for the health team in following up the success of treatment. In this study, an expert system was used to diagnose patients with potential Drop Out tuberculosis by using a Case Based Reasoning model to see patients with potential Droup Out. For variable names used are pulmonary smear patients (+), new patients, pulmonary smear (-) / ro (+), new patients, extra pulmonary, relapsed patients, re-treatment, default patients, re-treatment patients, failed patients and others -other. The last detection process is taken from the highest value obtained in the diagnosis of all the symptoms that have been witnessed. Based on the results of the application of the Expert System on Potential Drop Out Tuberculosis Patients at Cut Meutia Hospital in North Aceh based on the case code 31 with a detection system for the AFB (+) Lung Patient with its detection symptoms, the patient coughs with phlegm for 2-3 weeks or more. the results of sputum examination, patients who have been treated with TB drugs less than 1 month and TB patients on sputum examination, patients who have been treated with TB drugs less than 1 month, TB patients stop the treatment and TB patients return to the facility health service facilities with the highest case value of 0.6111 of all detection systems that have been tested.Keywords: Expert system,  CBS, TB
ANALISIS SENTIMEN ULASAN GAME HARRY POTTER: HOGWARTS MYSTERY PADA SITUS GOOGLE PLAY MENGGUNAKAN NAÏVE BAYES CLASSIFIER Rahman, Puti Ayu Andhini; Ermatita, Ermatita; Irmanda, Helena Nurramdhani
SEINASI-KESI Vol 4, No 1 (2021): SEINASI-KESI 2021
Publisher : Fakultas Ilmu Komputer UPN Veteran Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Salah satu dampak dari pandemi Covid-19 adalah semua orang harus berada di rumah untuk meminimalisir penyebaran virus dan mereka merasa jenuh karena taman rekreasi atau bermain tutup, untuk menghilangkan rasa jenuh tersebut banyak dari mereka yang mencari hiburan melalui permainan video baik online maupun offline. Harry Potter: Hogwarts Mystery merupakan permainan RPG bermain peran yang berlatarkan cerita Harry Potter berdasarkan seri novel JK Rowling yang dikembangkan oleh Jam City dengan lisensi dari Portkey Games. Ulasan pengguna merupakan salah satu hal penting untuk dijadikan pertimbangan developer. Untuk memantau ulasan tersebut, metode yang dilakukan pada penelitian ini menyangkut kegiatan pengumpulan data menggunakaan web scrapper, melalui preprocessing data, pembobotan kata, dan teknik yang digunakan untuk klasifikasi sentimen menggunakan Naïve Bayes Classifier yang dapat mengkategorikan ulasan pengguna tanpa harus melakukannya secara manual. Tujuan dari penelitian ini adalah untuk melakukan klasifikasi terhadap ulasan tersebut dan mendapatkan informasi yang berguna bagi pihak yang berkepentingan.