Naive Bayes for Thesis Labeling

The thesis preparation in the Department of Informatics Universitas Ahmad Dahlan is divided into two areas of interest, namely Intelligent Systems and Software and Data Engineering. Existing thesis title data is only used as an archive and has never been processed or classified to determine the trend of thesis topics based on student interest each year. The stages include data collection, the data is divided into two parts (training data and test data), manual labeling of training data, text preprocessing, and classification using Naive Bayes. The results show the trend of thesis title taking from 2013 to 2018 shows the thesis trend in the field of Intelligent Systems and Software. Accuracy testing uses Confusion Matrix and K-Fold Cross Validation with a k value is 10, has a value of 94.60%, precision of 97.30%, and a recall of 85.70%.

Long Short Term Memory on Bitcoin Price Forecasting

In modern times, many people rely on sophisticated technology to meet their needs. Already many technologies today can replace the role and function of society in the field of investment. There are many ways to fulfill the lives of these people, such as Bitcoin investment. Bitcoin is a digital asset that only exists in digital form by means of peer-to-peer work. To maximize profits, it is necessary to forecast Bitcoin prices when it will go up or down. This study tries to address the changes in Bitcoin prices whether to go up or down the next day with an artificial neural network model. The editor used in this study is the LSTM method. The data used is the Bitcoin blockchain data, namely time-series data in a one-day period from 1 January 2018 to 31 May 2019. Obtained forecasting results in June 2019 for Bitcoin to rise slowly and an accuracy value of 97.5% based on MAPE with the first day worth $8901.50.

Mobile Dictionary For Hitu Ethnic Language

Indonesians have many diverse ethnic groups. Each tribe has different traditions and cultures. As with language, each tribe has a different local language to communicate and interact with their community and environment. Hitu is one of the villages (Negeri) on Ambon Island, Central Maluku Regency. People in Tanah Hitu communicate every day using the Hitu language. Hituese is one of the local languages in Indonesia. The Hitu Country dictionary application is a mobile-based application that can make it easier for the people of Hitu Village to find translations from Hitu - Indonesian - English or vice versa. This is because many foreign and domestic tourists visit Hitu Village. People in Hitu can certainly speak Indonesian, but not all are able to speak it fluently. Therefore, this dictionary was created to facilitate the community in Hitu. This mobile application can display vocabulary translations that can be searched from Hitu to Indonesian and Indonesian to Hitu or vice versa. This dictionary will use Android-based mobile devices and will also take advantage of current developments that are very sophisticated. The Hitu dictionary application will be easy to carry anywhere and can be accessed every day.

Malware Static Analysis on Microsoft Macro Attack

In the 21st century, technology is increasing rapidly, the increase in technology is the potential for cyber attacks on today's technological infrastructure. Malware that is designed to damage computer systems without the owner's knowledge at a considerable cost becomes a cyber crime. This macro malware analysis is to study the code and behavior of malware when run on an operating system. To analyze this malware, this study uses a static analysis method by analyzing malware without running the program.

Live Forensics on GPS inactive Smartphone

Google is known to still track the user's location despite the GPS settings and location history in smartphone settings has been turned off by the user. This requires special handling to prove the location on smartphones with inactive GPS and view its Location History previously used by user. The research investigates if Google is still recording its user data location. Live Forensic requires data from the running system or volatile data which is usually found in Random Access Memory (RAM) or transit on the network. Investigations are carried out using a Google account with a method used by live forensics to obtain results from the location history. Smartphones have been checked manually through data backup through custom recovery that has been installed. When checking the backup filesystem, turned out that no location data is stored. Therefore, researchers conducted an analysis on the Google Account which was analyzed using a forensic tool to analyze cloud services to obtain location data results. The results of the analysis carried out obtained a similarity in location from 8-days investigations. Google can still find the location of smartphones with GPS disabled, but the location results are not accurate. Google can store user location data via cellular networks, Wi-Fi, and sensors to help estimate the user's location. The process of extracting the results from the google maps log using a Google account will be analyzed using the Elcomsoft Cloud eXplorer and Oxygen Forensic Cloud Extractor so that the log location results are still available by Google.

Simulasi Aplikasi Real Time Route Selection berbasis Wireless Sensor Network di Kota Makassar

Seiring dengan pertumbuhan kota yang secara di dunia melebihi 50%, persoalan kota menjadi lebih rumit dan kompleks, salah satunya adalah kemacetan yang dapat disebabkan oleh banjir serta pertumbuhan kendaraan yang melebihi luas jalanan. Hal ini juga terjadi di Kota Makassar sebagai ibu kota provinsi Sulawesi Selatan. Saat ini sering terjadi kemacetan lalu lintas di beberapa ruas jalan di Kota Makassar, terutama pada saat peak hours. Data terakhir Juni 2017 menunjukkan jumlah kendaraan di kota Makassar adalah sebanyak 1.425.635 dimana terjadi kenaikan lebih dari 100% dibandingkan tahun 2007. Pada penelitian ini akan dirancang sebuah aplikasi real time route selection dengan memanfaatkan Wireless Sensor Network sebagai penyedia data traffic condition pada ruas jalan. Algoritma pencarian rute terpendek yang digunakan adalah algoritma Dijkstra serta algoritma Floyd-Warshall yang dikombinasikan dengan fungsi skalar Chebycheff. Hasil penelitian menunjukkan bahwa rute yang dihasilkan oleh kedua algoritma sama tetapi algoritma Dijkstra memiliki waktu pemrosesan lebih cepat dengan rerata 7,45 ms. Kelemahan fungsi skalar Chebycheff adalah proses update jarak antar node yang dinamis bergantung pada perubahan kondisi lalu lintas. Hal ini dapat diatasi dengan menggunakan penggunaan metode inferensi lain untuk kriteria kondisi lalu lintas, seperti fuzzy logic maupun metode Multi Criteria Decision Making.

Kendali Linierisasi Umpan Balik pada Sistem Pendulum Terbalik

Pada penelitian ini diterapkan pengendali linierisasi umpan balik pada sistem pendulum terbalik yang memiliki karakteristik tidak linear dan tidak stabil. Berdasarkan pada pengujian, pengendali yang diusulkan mampu untuk membuat sistem mengikuti sinyal referensi undak dan sinus. Nilai respons sistem untuk sinyal referensi undak adalah waktu naik sebesar 4,0386; waktu kestabilan sebesar 3,4656 dan overshoot sebesar 0 persen. Oleh karena itu dapat disimpulkan bahwa pengendali umpan balik linierisasi mampu untuk mengendalikan sistem pendulum terbalik mengikuti sinyal referensi.In this study, a feedback linearization controller was applied to an inverted pendulum system that has non-linear and unstable characteristics. Based on the test, the proposed controller is able to make the system follow step and sine reference signals. The system response value for the step reference signal is an increment time of 4.0386; the time of stability was 3,4656 and overshoot was 0 percent. Therefore, it can be concluded that the linearized feedback controller is able to control the inverted pendulum system following the reference signal.

Single Exponential Smoothing-Multilayer Perceptron Untuk Peramalan Pengunjung Unik Jurnal Elektronik

Jumlah kunjungan rerata pengunjung unik per hari pada jurnal elektronik menunjukkan bahwa hasil terbitan karya ilmiah website tersebut menarik. Sehingga jumlah pengunjung unik dijadikan indikator penting dalam mengukur keberhasilan sebuah jurnal elektronik untuk memenuhi perluasan, penyebaran dan percepatan sistem akreditasi jurnal. Pengunjung Unik merupakan jumlah pengunjung per Internet Address (IP) yang mengakses sebuah jurnal elektronik dalam kurun waktu tertentu. Terdapat beberapa metode yang biasa digunakan untuk peramalan, diantaranya adalah Multilayer Perceptron (MLP).  Kualitas data berpengaruh besar dalam membangun model MLP yang baik, karena sukses tidaknya permodelan pada MLP sangat dipengaruhi oleh data input. Salah satu cara untuk meningkatkan kualitas data adalah dengan melakukan smoothing pada data tersebut. Pada penelitian ini digunkan metode peramalan Multilayer Perceptron berdasarkan penelitian sebelumnya dengan kombinasi data training dan testing 80%-20% dengan asitektur 2-1-1 dan learning rate 0,4. Selanjutnya untuk meningkatkan kualitas data dilakukan smoothing dengan menerapkan metode Single Exponential Smoothing. Dari penelitian yang dilakukan diperoleh hasil terbaik menggunakan alpha 0.9 dengan hasil akurasi MSE 94.02% dan RMSE 75.54% dengan lama waktu eksekusi 580,27 detik. The number of visits by the average unique visitor per day on electronic journals shows that the published scientific papers on the website are interesting. So that the number of unique visitors is used as an important indicator in measuring the success of an electronic journal to meet the expansion, dissemination and acceleration of the journal accreditation system. Unique Visitors is the number of visitors per Internet Address (IP) who access an electronic journal within a certain period of time. There are several methods commonly used for forecasting, including the Multilayer Perceptron (MLP). Data quality has a big influence in building a good MLP model, because the success or failure of modeling in MLP is greatly influenced by the input data. One way to improve data quality is by smoothing the data. In this study, the Multilayer Perceptron forecasting method was used based on previous research with a combination of training data and testing 80% -20% with a 2-1-1 architecture and a learning rate of 0.4. Furthermore, to improve data quality, smoothing is done by applying the Single Exponential Smoothing method. From the research conducted, the best results were obtained using alpha 0.9 with MSE accuracy of 94.02% and RMSE 75.54% with a long execution time of 580.27 seconds.

Enterprise Architecture Planning untuk Pengelolaan Masjid Muhammadiyah

Majelis Tabligh merupakan unsur pembantu pimpinan yang ditunjuk sebagai penanggungjawab lapangan dalam pengelolaan masjid Muhammadiyah. Sampai tulisan ini hadir, data-data berkait pengelolaan masjid masih belum ada, seperti jumlah masjid se-DIY, kondisi masjid, pengelolaan jamaah, pengelolaan keuangan masjid, dan lainnya. Pengawasan terhadap masjid-masjid Muhammadiyah juga masih belum bisa dilakukan, maka perlu dukungan sistem informasi/teknologi informasi (SI/TI) yang terintegrasi. Enterprise Architecture adalah gambaran hubungan logis yang terintegrasi antara visi misi, proses bisnis, system informasi, dan infrastruktur organisasi. Enterprise Architecture Planning (EAP) merupakan suatu metode atau kerangka acuan untuk membangun suatu arsitektur informasi yang mengacu pada tujuan atau kebutuhan bisnis organisasi yaitu arsitektur bisnis, arsitektur data, arsitektur aplikasi, arsitektur teknologi. Pada penelitian ini menghasilkan suatu arsitektur bisnis yang teridentifikasi berupa area fungsi, fungsi bisnis, dan proses bisnis Majelis Tabligh. Arsitektur data berupa 10 entitas data. Arsitektur aplikasi berupa 18 aplikasi. Arsitektur Teknologi yang teridentifikasi salahsatunya menggunakan platform web dan Mobile. Selain itu, didapatkan juga portofolio aplikasi dalam bentuk Mc Farlan Grid. The Majelis Tabligh is an auxiliary element to the leadership who is appointed to be in charge of the field in the management of the Muhammadiyah mosque. Until now, there are still no data related to mosque management, such as the number of mosques in Yogyakarta, the condition of mosques, management of congregations, mosque financial management, and others. Supervision of Muhammadiyah mosques is still not possible, so it is necessary to support an information system/information technology (IS / IT) system. Enterprise Architecture is a description of the integrated logical relationship between the vision and mission, business processes, information systems, and organizational infrastructure. Enterprise Architecture Planning (EAP) is a method or frame of reference for building an information architecture that refers to the goals or business needs of the organization, namely business architecture, data architecture, application architecture, technology architecture. This research resulted in an identified business architecture in the form of function areas, business functions, and the Majelis Tabligh business process. The data architecture consists of 10 data entities. The application architecture is in the form of 18 applications. One of the identified technology architectures uses web and mobile platforms. In addition, you can also get a portfolio of applications in the form of the Mc Farlan Grid.

Pelabelan Kelas Kata Bahasa Jawa Menggunakan Hidden Markov Model

Part of Speech Tagging atau POS Tagging adalah proses memberikan label pada setiap kata dalam sebuah kalimat secara otomatis. Penelitian ini menggunakan algoritma Hidden Markov Model (HMM) untuk proses POS Tagging. Perlakuan untuk unknown words menggunakan Most Probable POS-Tag. Dataset yang digunakan berupa 10 cerita pendek berbahasa Jawa terdiri dari 10.180 kata yang telah diberikan tagsetBahasa Jawa. Pada penelitian ini proses POS Tagging menggunakan dua skenario. Skenario pertama yaitu menggunakan algoritma Hidden Markov Model (HMM) tanpa menggunakan perlakuan untuk unknown words. Skenario yang kedua menggunakan HMM dan Most Probable POS-Tag untuk perlakuan unknown words. Hasil menunjukan skenario pertama menghasilkan akurasi sebesar 45.5% dan skenario kedua menghasilkan akurasi sebesar 70.78%. Most Probable POS-Tag dapat meningkatkan akurasi pada POS Tagging tetapi tidak selalu menunjukan hasil yang benar dalam pemberian label. Most Probable POS-Tag dapat menghilangkan probabilitas bernilai Nol dari POS Tagging Hidden Markov Model. Hasil penelitian ini menunjukan bahwa POS Tagging dengan menggunakan Hidden Markov Model dipengaruhi oleh perlakuan terhadap unknown words, perbendaharaan kata dan hubungan label kata pada dataset.  Part of Speech Tagging or POS Tagging is the process of automatically giving labels to each word in a sentence. This study uses the Hidden Markov Model (HMM) algorithm for the POS Tagging process. Treatment for unknown words uses the Most Probable POS-Tag. The dataset used is in the form of 10 short stories in Javanese consisting of 10,180 words which have been given the Javanese tagset. In this study, the POS Tagging process uses two scenarios. The first scenario is using the Hidden Markov Model (HMM) algorithm without using treatment for unknown words. The second scenario uses HMM and Most Probable POS-Tag for treatment of unknown words. The results show that the first scenario produces an accuracy of 45.5% and the second scenario produces an accuracy of 70.78%. Most Probable POS-Tag can improve accuracy in POS Tagging but does not always produce correct labels. Most Probable POS-Tag can remove zero-value probability from POS Tagging Hidden Markov Model. The results of this study indicate that POS Tagging using the Hidden Markov Model is influenced by the treatment of unknown words, vocabulary and word label relationships in the dataset.