Articles
201 Documents
Search results for
, issue
"Vol 2 No 12 (2018): Desember 2018"
:
201 Documents
clear
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)
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.
Rancang Bangun Sistem Pengenalan Rambu Petunjuk Arah Berbasis Raspberry Pi Menggunakan Metode OCR (Optical Character Recognition)
Rando Rando;
Hurriyatul Fitriyah;
Wijaya Kurniawan
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 (801.155 KB)
The road sign is one of the means that provides guidance or information to the driver or other road users, about the direction to be taken or the location of the city to go complete with the name and direction of where it is located. Road sign are required so the rider focuses on the road when driving. So that made the system of introduction of raspberry pi-based direction guidance by using OCR (Optical Character Recognition) method, made image processing road sign directions so that riders focus on the road then the information in the form of sound Then the system will separate colors other than green because of the color of the signage signs colored directions green, then the system will look for a box-shaped image, then processing the character of letters and arrows direction, then change the letters that have been detected in the previous process and then recognized to sound. From the test obtained the highest value of 100% and the lowest 0.13% of the characters detected on the signs of direction. The minimum time to execute the image to sound is 4.7 seconds, maximum of 8.02 seconds and 6.402 seconds on average. Based on the results of the research can be concluded that Optical Character Recognition proved able to recognize the detected image on the train data.
Identifikasi Ujaran Kebencian Pada Facebook Dengan Metode Ensemble Feature Dan Support Vector Machine
Aditya Kresna Bayu Arda Putra;
Mochammad Ali Fauzi;
Budi Darma Setiawan;
Eti Setiawati
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 (418.235 KB)
In the beginning, social media is used for socializing and interacting with other people. One of the most used social media for socializing is Facebook, with users amounting to over hundred million people around the world. Nowadays, on Facebook, its often found there's hate speech writing being shared at massive pace. Of course an assistance from language expert is a must for identifying hatespeech on Facebook because there's not yet an automatic system that can identify a hatespeech. The system in this research are made using Ensemble feature and Support Vector Machine. Ensemble feature is used for combining some of the feature extracted from each writing to ease the process of identifying a hatespeech. Support Vector Machine then used to identifying a hatespeech from a writing based on feature that are combined using ensemble feature. According to the result of testing, we acquired a 70% accuracy for the system so we can conclude that ensemble feature and support vector machine is good to use for identifying hatespeech on social media Facebook.
Implementasi Sistem Monitoring Pada Rumah Jamur Menggunakan Jaringan Nirkabel Berbasis Protokol Komunikasi Message Queuing Telemetry Transport (MQTT)
Nugraha Pangestu;
Rizal Maulana;
Rakhmadhany Primananda
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 (516.855 KB)
Nowadays, mushroom cultivations are growing rapidly in the society. Mushrooms have been becoming the popular groceries because of it's affordable price for all people. In it's growth process, Mushroom will need two main factors in order to grow properly; which are : temperature and humidity of the room. These two factors need to be controlled continuously to fit the ideal value limit. The ideal temperature for mushroom growth ranges between 25-30 â—‹C and the air humidity ranges between 60-90%. To overcome these problems, we proposed a system that can continuosly monitor temperature and humidity of mushroom house using MQTT protocol. To detect temperature and humidity, DHT-22 sensor is used. The sensor data will be processed by Arduino microcontroller and then sent using ESP8266-01 Wi-fi module which connected through the Wi-fi network to the Thingsboard server using MQTT protocol. Finally the data will be displayed in the form of charts and graphs. We evaluate the system using the delay test by calculating the time required by each sensor to process and read the data until the data is sent and displayed on the Thingsboard server. The results of delay test have confirmed that the system takes an average of 1.20 seconds in a single delivery of data to the Thingsboard server.
Sistem Deteksi Kerusakan Mesin Pada Sepeda Motor Menggunakan Naive Bayes - Certainty Factor
Alfan Nazala Putra;
Nurul Hidayat;
Suprapto Suprapto
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 (1020.7 KB)
Nowadays, motorcycle is no longer a luxury item for most people. Almost the whole community at least had motorcycle. Motorcycle became one of the main means of transportation which more dynamic and faster compared to other means of transport, and this is provable by the large number of motorcyclists compared to another user of transportation means on the road. Not surprising that motorcycle is the largest cause for accident in traffic. One of the causes of the accident on a motorcycle is from motorcycle engines. But as most motorcyclist are still much less savvy about the damage to their motorcycle engines because there are various kind of failure. A method of classification can be implemented into the software to know which part of motorcycle engines that get damaged or failure. One example is the Naive Bayes. Naive Bayes Classifier is a simple probability classification based on Bayes theorem which has Bayes inference in general, especially with a strong independence of assumptions. The theory of certainty using a value called the Certainty Factor (CF) to assume a degree of confidence to a data. The variable used in this study is a list of symptoms and damage to motorcycle engines. Highest accuracy resulting in this research is 90%.
Implementasi Sistem Pendeteksi Rogue Access Point Dengan Metode Perhitungan Nilai Round Trip Time
Fahmi Syahrulah;
Adhitya Bhawiyuga;
Mahendra Data
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 (259.878 KB)
With the increasing development of wireless technology, which provoke the cyberattack using wireless technology increase as well, this would make user data be put to risk. One of the cyberattack that exploit wireless technology is Rogue Access Point. To prevent this, RAP detection system was made with comparing round trip time method. This system sent a DNS lookup to get round trip time. RAP would return much longer round trip time value than the legitimate access point, this happen because the added hop that was passed during the trip to RAP. To minimalize the impact from unstable network that would affect the accuracy of detection system, filtering mechanism is used to filter round trip time value that is deemed as an anomaly. Testing will be done to the two similar access point which one of them is the RAP. Round trip time value from both access point will be compared, which will be used as parameter to determine RAP. In this research the accuracy detection would be never less than 90%.
Klasifikasi Hoax Pada Berita Kesehatan Berbahasa Indonesia Dengan Menggunakan Metode Modified K-Nearest Neighbor
Andre Rino Prasetyo;
Indriati Indriati;
Putra Pandu Adikara
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 (926.202 KB)
News is a source of information about current events which can be found in newspapers, television, the internet and other media. Currently the news that is disseminated often without writing the source clearly, especially the type of news about health, it can lead to misinterpretation because the news is not necessarily true or wrong so it takes a smart system to classify health news is whether included in the category of hoax or fact. The hoax classification process use several stages ranging from preprocessing consisting of tokenisasi and filtering. Continued with word-weighting process and cosine similarity to classification process using Modified K-Nearest Neighbor method. The results obtained based on the implementation and testing resulted in the best value of k amounted to 4, precision of 0,83 recall of 0,75 f-measure of 0,79 and the accuracy of 75%. The test results obtained because the health news content used is still too common, many non-standard words and the determination of k-values ​​used are very influential on whether or not the process of classification of health news documents.
Implementasi Metode F-KNN (Fuzzy K-Nearest Neighbor) Untuk Diagnosis Penyakit Anjing
Dizka Maryam Febri Shanti;
Nurul Hidayat;
Randy Cahya Wihandika
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 (376.438 KB)
Dogs are one of the favorite animals used as pets. When petting and playing with dogs, oxytocin, stress-related and relieved hormones are released, helping to lower blood pressure as well as cortisol levels. Although dog maintenance has many benefits, the owners should be careful in caring for their dogs. Not a few dogs are attacked by various diseases caused by viruses, protozoa, bacteria and parasites. Dogs who are sick if not immediately get treatment and treatment have the risk of transmitting to dogs and other animals or even to humans. The method used Fuzzy K-Nearest Neighbor. (FK-NN) is a variant of K-Nearest Neighbor (K-NN) method with fuzzy technique. The FK-NN method assigns a class membership value to the sample vector instead of placing the vector in a particular class. FK-NN can be implemented for the diagnosis of diseases in dogs by several stages: calculating the distance between the train data and the test data, taking the smallest distance between the train data and the test data as much as K, Fuzzification and Defuzzification, Class with the highest defuzzification value used as the class for the result classification. The value of K affects the accuracy of the system where the higher the value of k then the tendency of accuracy will decrease. The highest accuracy obtained from the test results is when K = 5 ie with a value of 98.67%.
Evaluasi Kebutuhan Fungsional Sistem Informasi Pelaporan Kegiatan Menggunakan Metode Quality Function Deployment di BNN Kota Malang
Sinta Kurniasari;
Mochamad Chandra Saputra;
Admaja Dwi Herlambang
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 (344.929 KB)
In the information system development process, information system required the presence of definition and determination of priority functional requirements. The purpose of the definition and determination of priorities is to help organizations found out where the functional requirements are indeed important and necessary. This research was conducted on the definition of the functional requirements and the determination of priorities by using the method of Quality Functional Deployment (QFD). QFD method used to discover the functional requirements of the system based on the desires of the customer. The matrix was used in this research is the matrix of House of Quality (HoQ). This research was conducted using qualitative methods to interview informants that add up to 8 people. From the result of preparation of HoQ produced a number of 24 VoC and 28 functional requirements, as well as of the calculation of the weighted importance rating with weights relation obtained priority value for the 28 functional needs.
Analisis Proses Bisnis Pada Toko Buku Galuh Menggunakan Business Process Improvement Framework
Adeyaksa Galuh Waluyo;
Ismiarta Aknuranda;
Nanang Yudi Setiawan
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 (474.184 KB)
Toko Buku Galuh is a company engaged in retail business. Toko Buku Galuh has stood since 1998, and has grown to the present. Vision Galuh Bookstore to be at the forefront of the retail business of books in Sidoarjo. Problems arise as the company grows larger that there is no written business process that becomes the benchmark in running the business, so that the decisions made can be changed based on the conditions leading to less effective and efficient business processes. For that required analysis of business processes using Business Process Improvement (BPI) Framework. The first phase of editing Toko Buku Galuh business process uses Business Process Model and Notation (BPMN), the second phase performs Process Validation, Time Validation and Resource Validation on business processes. Then determine the streamlining used to improve the effectiveness and efficiency of business processes. From the results of the proposed analysis obtained increased time on sale transactions 64.37%, buyer transactions 77,9%, price tagging 30%, and edit items data 25.9%.