Claim Missing Document
Check
Articles

Found 27 Documents
Search

PERPADUAN COMBINED SAMPLING DAN ENSEMBLE OF SUPPORT VECTOR MACHINE (ENSVM) UNTUK MENANGANI KASUS CHURN PREDICTION PERUSAHAAN TELEKOMUNIKASI Marbun, Fernandy; Baizal, Abdurahman; Bijaksana, Moch Arif
JUTI: Jurnal Ilmiah Teknologi Informasi Vol 8, No 2, Juli 2010
Publisher : Department of Informatics, Institut Teknologi Sepuluh Nopember

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (11932.256 KB) | DOI: 10.12962/j24068535.v8i2.a316

Abstract

Churn prediction adalah suatu cara untuk memprediksi pelanggan yang berpotensial untuk churn. Data mining khususnya klasifikasi tampaknya dapat menjadi alternatif solusi dalam membuat model churn prediction yang akurat. Namun hasil klasifikasi menjadi tidak akurat disebabkan karena data churn bersifat imbalance. Kelas data menjadi tidak stabil karena data akan lebih condong ke bagian data yang memiliki komposisi data yang lebih besar. Salah satu cara untuk menangani permasalahan ini adalah dengan memodifikasi dataset yang digunakan atau yang lebih dikenal dengan metode resampling. Teknik resampling ini meliputi over-sampling, under-sampling, dan combined-sampling. Metode Ensemble of SVM (EnSVM) diharapkan dapat meminimalisir kesalahan klasifikasi kelas mayor dan minor yang dihasilkan oleh classifier SVM tunggal. Dalam penelitian ini akan dicoba untuk memadukan combined sampling dan EnSVM untuk churn predicition. Pengujian dilakukan dengan membandingkan hasil klasifikasi CombinedSampling-EnSVM dengan SMOTE-SVM (perpaduan oversamping-SVM) dan pure-SVM. Hasil pengujian menunjukkan bahwa metode CombinedSampling-EnSVM secara umum hanya mampu menghasilkan performansi Gini Index yang lebih baik daripada metode SMOTE-SVM dan tanpa resampling (pure-SVM).
Analisis dan Klasifikasi Opini pada Porduct Review Menggunakan Metode Semi-Supervised annisa Imadi Puti; Warih Maharani; Mochammad Arif Bijaksana
Indonesia Symposium on Computing Indonesia Symposium on Computing (IndoSC) 2016
Publisher : Indonesia Symposium on Computing

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

Abstract

Review produk merupakan informasi penting bagi konsumen dan produsen. Bagi konsumen, review sering dijadikan sebagai referensi dan acuan untuk mengambil keputusan. Jumlah review produk yang banyak membuat isi review tidak dapat disimpulkan dengan cepat dan tepat. Untuk mengatasi masalah tersebut, diperlukan suatu sistem yang secara otomatis dapat mengidentifikasi fitur-fitur dan review dan mengklasifikasikannya ke dalam polaritas positif atau negatif. Penilitian tugas akhir ini dilakukan untuk menganalisis klasifikasi dari review produk. Sebelum memasuki analisis klasifikasi, penelitian dimulai dari proses ekstraksi fitur menggunakan metode type dependency parser, identifikasi noun phrase parser dan AER. Fitur hasil ekstraksi kemudian dilihat polaritas opininya menggunakan metode semi-supervised dengan melakukan pembangunan graph berbasis lexicon berisi kata-kata opini positif-negatif yang dikembangkan sinonimnya.
ANALISIS SEMANTIS KALIMAT BERBAHASA INDONESIA DENGAN METODE BERBASIS KONSEP Indrawati, Nur; Triawati, Candra; Saputro3,, Widyanto Adi; Bijaksana, Moch. Arif
Masyarakat Telematika Dan Informasi : Jurnal Penelitian Teknologi Informasi dan Komunikasi Vol 5, No 2 (2014): Masyarakat Telematika Dan Informasi : Jurnal Penelitian Teknologi Informasi dan
Publisher : Kementerian Komunikasi dan Informatika R.I.

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

Abstract

One of semantic approach in word or phrase separation for determining topic of a document is a concept-based mining model. This approach works based on frequency of occurrence of the word by heeding semantic aspects, so the results are more accurate than syntax analyzing. Concept-based mining model arises from the need for understanding and manipulating the text to get specific information from text document using computer technology, by considering the aspects of syntactic and semantic.Semantic role labeling that based on case grammar is used to obtain semantic role of each word or phrase in a document, whereas the label itself are taken from PropBank. Concept-weighting in this research used two concept-based methods, the graphical concept-based where each concept is represented in a graph rise with different values for each levels, and the statistical concept-based that analyze the concept statistically which based on the frequency of occurrence. The concept with the highest weight value will represent the topic of the document. The test results that acquired by using test data which is news article that used Bahasa, shows that the number of attributes that represent the document on conceptbased mining models approach are affecting system performance,but the given effects are different between classification and clustering. Identifying semantic roles provide the level of semantic analysis to solve human language processing problems. Further research is needed to get higher quality of concept.
Name Disambiguation Analysis Using the Word Sense Disambiguation Method in Hadith Prasetio, Ageng; Bijaksana, Mochammad Arif; Suryani, Arie Ardiyanti
Jurnal Pendidikan Informatika (EDUMATIC) Vol 4, No 2 (2020): Edumatic : Jurnal Pendidikan Informatika
Publisher : Universitas Hamzanwadi

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

Abstract

Name disambiguation is the problem solving process to find similar names in sentences. The ambiguity of names can be found in hadith of Sahih Bukhari, names "Abdullah bin Amru" in hadiths no 27 and “Abdullah bin Amru” in hadith no 58, These names are the same, but there is no proof they are the same person. This problem is the early indication of ambiguity of name in the hadith. Based in this problem, this research aims to find name disambiguation of hadith narrators with classification by considering the perawi chain. To solved this problem the authors used Word Sense Disambiguation (WSD), WSD is a process to assign the same meaning from the sentences, based on the context in which the word appears. To classify several names in the hadith, the authors used KNN algorithm, by combining the WSD and KNN method can reduce the ambiguity of names in hadith. The data used in this study came from the hadith of Sahih Bukhori through the pre-processing stage. After conducting the research showed a collection of hadith numbers with the same name prediction with an accuracy of 99% at k = 1. Thus, this method can be used for name disambiguation.
Building Synonym Sets for English WordNet with Robust Clustering using Links Method Suryaningsih, Sarah; Bijaksana, Moch Arif; Astuti, Widi
Jurnal Pendidikan Informatika (EDUMATIC) Vol 4, No 1 (2020): Edumatic : Jurnal Pendidikan Informatika
Publisher : Universitas Hamzanwadi

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

Abstract

English WordNet is an important synonym set to present the similarity of meanings between words. Synonym Set is built using Oxford Thesaurus which is accessed through lexico.com, which is a part of the lexical database that will be used. After using the extraction process through Oxford Thesaurus it will produce a synonym set with the same meaning between words. The difference between WordNet and ordinary dictionaries is that the word is interconnected with other words. One method employed for this approach is Robust Clustering Using Links method, which is similarity values and synonym sets that have been created to be used to build a lexical database. Therefore the main purpose of the development of the English WordNet is to produce an accurate synonym set using clustering techniques. The evaluation calculation will use the F-measure method and will use the gold standard for the calculation method. With the ROCK method, there is an increase in accuracy output from dataset input. Building the English wordnet is to improve words that can be used to help research and development of other language wordnets with role models using more accurate English wordnets. And the use of ROCK method there is an increase in the accuracy upon results of the development of English wordnet compared to the previous method, which is using hierarchical clustering. The outcome of this study resulted in improved accuracy so that the ROCK method is one of the good methods used in the development of the English wordnet.
Developing Set of Word Senses of Vocabulary in Al-Qur’an Aqila, Neca; Bijaksana, Mochammad Arif
Jurnal Pendidikan Informatika (EDUMATIC) Vol 4, No 1 (2020): Edumatic : Jurnal Pendidikan Informatika
Publisher : Universitas Hamzanwadi

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

Abstract

Al-Qur’an has become the guideline for all Muslims in the world, which makes many Muslims are eager to understand its contents. Nevertheless, Al-Qur’an consist of many words that have more than one meaning, which represent certain difficulties while understanding the meaning itself. As example, the word أَزْوَاجًا has two equivalents as it might be translated either "jodoh" (“mate”) in Surah An-Nahl (16: 72: 6) and "golongan-golongan" (“groups”) in Surah Al-Hijr (15: 88: 8). This case is known as a word sense, a word that has more than one meaning. This research aims to construct the word sense as a set of vocabulary, in order to simplify the vocabulary meaning in Al-Qur’an itself. The data set used in this research is nouns from Al-Qur’an which have been translated into Bahasa. In order to construct the set of word sense, this research grouped the words using Hierarchical Clustering method. The total set of the word senses found is 34 nouns, which contains diverse translation. The F measure from evaluation of this research resulted in an accuracy of 65.85%. The result was obtained based on the conformity between the results of the word senses set by the system and by the linguists. The outcome of this research is, accuracy is low, due to the type and the number data used is limited.
Dependency Parsing for Arabic Quran using Easy-First Parsing Algorithm Hafsa, Alfiya El; Bijaksana, Mochammad Arif; Huda, Arief Fatchul
Jurnal Pendidikan Informatika (EDUMATIC) Vol 4, No 2 (2020): Edumatic : Jurnal Pendidikan Informatika
Publisher : Universitas Hamzanwadi

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

Abstract

Arabic is the main language of Al-Quran. Nowadays, many people are studying the Language of Al-Quran, called Quran Arabic. For the beginners, it is important for them to understand the syntactic relationship in a sentence found in the Qur'an. If they do not understand enough, the interpretation will be different and wrong. It will turn into dangerous because Al-Quran is a source of guidance for Muslims’ life. Dependency parsing is very important for linguistic research, especially for rich languages such as the Arabic Language. This study aims to build dependency parsing, in order to make it easier to get to understand syntactic relationship information in sentences. This study uses a parsing method called deterministic parsing, which the method used is shift-reduce parsing with the Easy-First parsing algorithm. The evaluation used labeled attachment score calculation. The score generated from the evaluation was 69.7, beforehand, the comparison both the system results and the gold standard have been done. 62 sentences found the correct head and relation in each word. The number of words found to be wrong is not more than 3 words in one sentence. Evaluation scores produced are not exorbitant due to the complicated tagset used and lacking test sentences.
Typo handling in searching of Quran verse based on phonetic similarities Purwita, Naila Iffah; Bijaksana, Moch Arif; Lhaksmana, Kemas Muslim; Naf’an, Muhammad Zidny
Register: Jurnal Ilmiah Teknologi Sistem Informasi Vol 6, No 2 (2020): July
Publisher : Information Systems - Universitas Pesantren Tinggi Darul Ulum

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26594/register.v6i2.2065

Abstract

The Quran search system is a search system that was built to make it easier for Indonesians to find a verse with text by Indonesian pronunciation, this is a solution for users who have difficulty writing or typing Arabic characters. Quran search system with phonetic similarity can make it easier for Indonesian Muslims to find a particular verse.  Lafzi was one of the systems that developed the search, then Lafzi was further developed under the name Lafzi+. The Lafzi+ system can handle searches with typo queries but there are still fewer variations regarding typing error types. In this research Lafzi++, an improvement from previous development to handle typographical error types was carried out by applying typo correction using the autocomplete method to correct incorrect queries and Damerau Levenshtein distance to calculate the edit distance, so that the system can provide query suggestions when a user mistypes a search, either in the form of substitution, insertion, deletion, or transposition. Users can also search easily because they use Latin characters according to pronunciation in Indonesian. Based on the evaluation results it is known that the system can be better developed, this can be seen from the accuracy value in each query that is tested can surpass the accuracy of the previous system, by getting the highest recall of 96.20% and the highest Mean Average Precision (MAP) reaching 90.69%. The Lafzi++ system can improve the previous system.
Pencarian Potongan Ayat Al-Qur'an dengan Perbedaan Bunyi pada Tanda Berhenti Berdasarkan Kemiripan Fonetis Naufal Rasyad; Moch. Arif Bijaksana; Kemas Muslim Lhaksmana
Jurnal Linguistik Komputasional Vol 2 No 2 (2019): Vol. 2, No. 2
Publisher : Indonesia Association of Computational Linguistics (INACL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1238.009 KB) | DOI: 10.26418/jlk.v2i2.25

Abstract

Al-Quran merupakan kitab suci utama bagi umat Islam yang ditulis menggunakan bahasa Arab. Seiring dengan perkembangan teknologi, telah dikembangkan sistem pencarian ayat Al-Qur’an berdasarkan kemiripan fonetis salah satunya adalah Lafzi. Namun untuk menangani perbedaan bunyi pada tanda berhenti di pertengahan ayat, sistem Lafzi belum bisa menanganinya dengan baik. Maka dari itu, dibutuhkan sistem yang dapat membantu pengguna dalam melakukan pencarian ayat Al-Quran, terutama untuk perbedaan bunyi pada tanda berhenti sehingga pencarian bisa menemukan kata yang berbeda pengucapan pada tanda berhenti. Berdasarkan permasalahan tersebut, dari sistem Lafzi, dilakukan pengembangan supaya dapat melakukan pencarian yang bisa menangani perbedaan bunyi pada tanda berhenti. Digunakan pengindeksan trigram untuk memperkirakan kecocokan string antara kueri dengan transliterasi ayat Al-Qur’an serta dibuat aturan pada input dengan huruf akhir ’T’ menjadi ’H’. Sistem yang sudah ada mendapatkan nilai recall sebesar 81% dan nilai MAP sebesar 65%. Sedangkanhasildaripenelitianinidiperolehnilairecallsebesar 100% dan nilai MAP sebesar 84%.
Building Monolingual Word Alignment For Indonesian Al-Quran Translation Galih Rizky Prabowo; Moch Arif Bijaksana
Jurnal Linguistik Komputasional Vol 1 No 2 (2018): Vol. 1, No. 2
Publisher : Indonesia Association of Computational Linguistics (INACL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (986.575 KB) | DOI: 10.26418/jlk.v1i2.11

Abstract