Garuda - Garba Rujukan Digital

Pengenalan Suara Paru-Paru dengan MFCC sebagai Ekstraksi Ciri dan Backpropagation sebagai Classifier Syafria, Fadhilah; Buono, Agus; Silalahi, Bib Paruhum
Jurnal Ilmu Komputer dan Agri-Informatika Vol 3, No 1 (2014)
Publisher : Departemen Ilmu Komputer IPB

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (568.648 KB)

Paru-paru merupakan organ vital manusia yang berperan dalam proses pernapasan. Jika paru-paru mengalami gangguan maka sistem pernapasan manusia juga akan mengalami gangguan yang bisa menyebabkan kecacatan bahkan kematian. Untuk mengevaluasi keadaan paru-paru dapat dilakukan dengan mendengarkan suara pernapasan dengan menggunakan stateskop. Teknik ini dikenal dengan teknik auskultasi. Teknik ini paling sering digunakan namun memiliki beberapa kelemahan yaitu suara paru-paru berada pada frekuensu rendah, masalah kebisingan lingkungan, kepekaan telinga, hasil analisa yang subjektif, dan pola suara yang hampir mirip. Karena faktor-faktor di atas kesalahan diagnosa bisa terjadi jika proses auskultasi tidak dilakukan dengan benar. Dalam penelitian ini, akan dibuat pengenalan suara paru-paru normal dan abnormal menggunakan Mel Frequency Cepstrum koefisien (MFCC) sebagai ekstraksi ciri dan Backpropagation sebagai classifier. Suara paru-paru akan dihitung Coeffisient Ceptral nya sebagai penciri dari masing-masing suara untuk selanjutnya dikenali dengan menggunakan Backpropagation. Metode yang diusulkan memberikan akurasi 93.97% untuk data latih dan 92.66% untuk data uji.Kata kunci: Backpropagation, MFCC, pengenalan suara paru-paru

Optimasi K-Means Clustering Menggunakan Particle Swarm Optimization pada Sistem Identifikasi Tumbuhan Obat Berbasis Citra Bisilisin, Franki Yusuf; Herdiyeni, Yeni; Silalahi, Bib Paruhum
Jurnal Ilmu Komputer dan Agri-Informatika Vol 3, No 1 (2014)
Publisher : Departemen Ilmu Komputer IPB

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1151.464 KB)

Teknologi identifikasi pada penelitian ini diperlukan untuk mempercepat proses identifikasi spesies tumbuhan obat berupa data citra digital. Penelitian ini membangun sistem identifikasi tumbuhan obat menggunakan teknik clustering. Teknik clustering digunakan untuk mengelompokkan data citra sesuai dengan spesies tumbuhan obat. Penelitian ini bertujuan melakukan optimasi k-means clustering menggunakan metode particle swarm optimization (PSO). Metode PSO digunakan untuk mengatasi kelemahan pada metode clustering tradisional yaitu pemilihan pusat cluster awal dan solusi lokal. Proses ekstraksi fitur menggunakan fuzzy local binary pattern (FLBP) untuk merepresentasikan tekstur dari citra. Implementasi program menggunakan bahasa pemrograman C++. Analisis clustering dilakukan untuk 30 spesies tumbuhan obat yang ada di Indonesia dengan jumlah 48 citra masing-masing spesies. Pengukuran kualitas clustering menggunakan nilai quantization error dan akurasi. Hasil yang diperoleh menunjukkan metode PSO mampu meningkatkan kinerja dari metode k-means clustering dalam proses identifikasi tumbuhan obat.Kata kunci: fuzzy local binary pattern, k-means clustering, particle swarm optimization, tumbuhan obat

ANALISIS PEMBENTUKAN POLA GRAF PADA KALIMAT BAHASA INDONESIA MENGGUNAKAN METODE KNOWLEDGE GRAPH Yusuf, Yasin; Nurdiati, Sri; Silalahi, Paruhum
Lingua Vol 10, No 1 (2014): January 2014
Publisher : Lingua

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

Knowledge graph adalah sebuah pendekatan baru untuk memahami bahasa alami. Metodeini memiliki 9 relasi biner dan 4 relasi frame. Analisis suatu kalimat dengan menggunakanknowledge graph membutuhkan aturan pemotongan kalimat (chunking). Aturan chunkingsudah ada pada struktur kalimat bahasa Inggris dan Cina, tetapi belum ada untuk strukturkalimat bahasa Indonesia.Tujuan dari penelitian ini adalah membentuk aturan chunkingpada struktur kalimat bahasa Indonesia dan membuat pola graf kalimat bahasaIndonesia.Tahapan penelitian ini adalah dimulai dengan studi literatur awal, pembuatanchunk indicator, pemotongan kalimat (chunking), pembuatan chunk graph, dan diakhiridengan kontruksi sentence graph. Hasil penelitian ini adalah aturan chunking kalimatbahasa Indonesia dengan indicator sebanyak 8, yaitu koma dan titik, kata ganti petunjuk,kata kerja bantu, kata depan, jump, kata-kata logika, jeda nafas, kata sambung. Selain itu,diperoleh pula pola graf kalimat bahasa Indonesia yang sekaligus menunjukkan arti(aspek semantik) dari kalimat yang dianalisis.Â This research aimed to construct chunking rule on Indonesian language sentencestructure and make pattern of Indonesian language sentence graph. It was done sinceknowledge graph is a new approach to understand natural language. This method has 9(nine) binary relation and 4 (four) frame relation. A sentence analysis using this approachneeds rule of sentence chunking, This research method was started from beginning ofliterary studies, chunk indicator constructing, sentence chunking, chunk graphconstructing, and sentence graph constructing. Result of this research was there was ruleof Indonesian language sentence chunking with 8 (eight) indicators such as periods, fullstops, demonstratives, auxiliary verbs, prepositions, jump, logical words, pauses,conjunctions. Besides that, it had also been achieved pattern of Indonesian languagegraph which gives meaning (semantic aspect) from analyzed sentences at once.

ANALISIS PEMBENTUKAN POLA GRAF PADA KALIMAT BAHASA INDONESIA MENGGUNAKAN METODE KNOWLEDGE GRAPH Yusuf, Yasin; Nurdiati, Sri; Silalahi, Paruhum
Lingua Vol 10, No 1 (2014): January 2014
Publisher : Lingua

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

Knowledge graph adalah sebuah pendekatan baru untuk memahami bahasa alami. Metodeini memiliki 9 relasi biner dan 4 relasi frame. Analisis suatu kalimat dengan menggunakanknowledge graph membutuhkan aturan pemotongan kalimat (chunking). Aturan chunkingsudah ada pada struktur kalimat bahasa Inggris dan Cina, tetapi belum ada untuk strukturkalimat bahasa Indonesia.Tujuan dari penelitian ini adalah membentuk aturan chunkingpada struktur kalimat bahasa Indonesia dan membuat pola graf kalimat bahasaIndonesia.Tahapan penelitian ini adalah dimulai dengan studi literatur awal, pembuatanchunk indicator, pemotongan kalimat (chunking), pembuatan chunk graph, dan diakhiridengan kontruksi sentence graph. Hasil penelitian ini adalah aturan chunking kalimatbahasa Indonesia dengan indicator sebanyak 8, yaitu koma dan titik, kata ganti petunjuk,kata kerja bantu, kata depan, jump, kata-kata logika, jeda nafas, kata sambung. Selain itu,diperoleh pula pola graf kalimat bahasa Indonesia yang sekaligus menunjukkan arti(aspek semantik) dari kalimat yang dianalisis.Â This research aimed to construct chunking rule on Indonesian language sentencestructure and make pattern of Indonesian language sentence graph. It was done sinceknowledge graph is a new approach to understand natural language. This method has 9(nine) binary relation and 4 (four) frame relation. A sentence analysis using this approachneeds rule of sentence chunking, This research method was started from beginning ofliterary studies, chunk indicator constructing, sentence chunking, chunk graphconstructing, and sentence graph constructing. Result of this research was there was ruleof Indonesian language sentence chunking with 8 (eight) indicators such as periods, fullstops, demonstratives, auxiliary verbs, prepositions, jump, logical words, pauses,conjunctions. Besides that, it had also been achieved pattern of Indonesian languagegraph which gives meaning (semantic aspect) from analyzed sentences at once.

Deteksi dan Identifikasi Pelaku Kecurangan Skema Pembagian Rahasia Linear Berbasis Skema Shamir Nur Ahzan, Zulkaidah; Guritman, Sugi; Silalahi, Bib Paruhum
Jurnal Karya Pendidikan Matematika Vol 7, No 1 (2020): Jurnal Karya Pendidikan Matematika Volume 7 Nomor 1 Tahun 2020
Publisher : Universitas Muhammadiyah Semarang

The method that can be used to maintain security of secret in the form of cryptographic keys is by using secret sharing scheme (SSS). This method is first proposed by Adi Shamir in 1979, where the proposed scheme is a (k, n) threshold scheme. Shamir scheme is a perfect scheme under the assumption that all shareholders present their original share. However, if there are dishonest shareholders who present faked shares then the honest shareholders get nothing but a faked secret. Secret sharing scheme based on linear scheme is a scheme that can detect and identify cheaters who submit faked shares at the secret reconstruction. Detectability of this scheme when and identifiability when under the assumption that all shareholders present their shares randomly. After conducting a security analysis of the proposed scheme, it is obtained that to succeed in attack with cheaters who work together to fool honest shareholders then a new polynomial g(x) such that g(1) = , g(2) = , …, g(k - 1) = and a new detector that has the same value as detector d are needed.

APLIKASI ZERO-ONE GOAL PROGRAMMING DALAM MASALAH PEMILIHAN PROYEK PEMASARAN Silalahi, Bib Paruhum; Pertiwi, Silviana Eka; Mayyani, Hidayatul; Aliatiningtyas, Nur
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 14 No 3 (2020): BAREKENG: Jurnal Ilmu Matematika dan Terapan
Publisher : MATHEMATIC DEPARTMENT, FACULTY OF MATHEMATICS AND NATURAL SCIENCES, UNIVERSITY OF PATTIMURA

Marketing management is an activity to plan and organize marketing activities in order to achieve organizational or company goals efficiently and effectively. Problems arise when there are several or many different projects that can be implemented as company marketing projects. These projects are usually categorized by several objectives. These goals can be complementary or contradictory. In operation, decision-makers are required to choose and determine the right project to achieve the target. In this paper, we discuss a programming model using the zero-one goal programming approach, a selection of marketing projects to meet many objectives and constraints, and then give examples of its implementation. Discussion and implementation include goal programming categories: nonpreemptive goal programming and preemptive goal programming

Modeling Text Independent Speaker Identification with Vector Quantization Syeiva Nurul Desylvia; Agus Buono; Bib Paruhum Silalahi
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 15, No 1: March 2017
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v15i1.4656

Speaker identification is one of the most important technology nowadays. Many fields such as bioinformatics and security are using speaker identification. Also, almost all electronic devices are using this technology too. Based on number of text, speaker identification divided into text dependent and text independent. On many fields, text independent is mostly used because number of text is unlimited. So, text independent is generally more challenging than text dependent. In this research, speaker identification text independent with Indonesian speaker data was modelled with Vector Quantization (VQ). In this research VQ with K-Means initialization was used. K-Means clustering also was used to initialize mean and Hierarchical Agglomerative Clustering was used to identify K value for VQ. The best VQ accuracy was 59.67% when k was 5. According to the result, Indonesian language could be modelled by VQ. This research can be developed using optimization method for VQ parameters such as Genetic Algorithm or Particle Swarm Optimization.

Kompresi Data Menggunakan Algoritme Huffman Julio Adisantoso; Danny Dimas Sulistio; Bib Paruhum Silalahi
Seminar Nasional Aplikasi Teknologi Informasi (SNATI) 2004
Publisher : Jurusan Teknik Informatika, Fakultas Teknologi Industri, Universitas Islam Indonesia

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

Text compression algorithms are normally defined in terms of a source alphabet of 8-bit ASCII codes. Huffman algorithm is the most popular methods of text compression. Thisresearch used static and adaptif Huffman algorithms to compress text data, and also compareit. Variation of character occurs will decrease compression ratio. Iteration time of staticHuffman algorithm for compress and decompress is faster than adaptif Huffman algorithm,but performance of adaptif Huffman algorithm is best.Keywords: text compression, static and adaptif huffman algorithm.

Use of Ant Colony Optimization Algorithm for Determining Traveling Salesman Problem Routes Bib Paruhum Silalahi; Nurul Fathiah; Prapto Tri Supriyo
Jurnal Matematika MANTIK Vol. 5 No. 2 (2019): Mathematics and Applied Mathematics
Publisher : Mathematics Department, Faculty of Science and Technology, UIN Sunan Ampel Surabaya

Ant Colony Optimization is one of the meta-heuristic methods used to solve combinatorial optimization problems that are quite difficult. Ant Colony Optimization algorithm is inspired by ant behavior in the real world to build the shortest path between food sources and their nests. Traveling Salesman Problem is a problem in optimization. Traveling Salesman Problem is a problem to find the minimum distance from the initial node to the whole node with each node must be visited exactly once and must return to the initial node. Traveling Salesman Problem is a non-deterministic polynomial-time complete problem. This research discusses the solution of the Traveling Salesman Problem using the Ant Colony Optimization algorithm and also using the exact algorithm. The results showed that the greater the size of the Traveling Salesman Problem case, the longer the execution time required. The results also showed that the execution times of the Ant Colony Optimization are much faster than the execution time of the exact method.

Konstruksi Persegi Ajaib dengan Entri Bilangan Bulat Ulil Albab Mu'min; Bib Paruhum Silalahi
JMPM: Jurnal Matematika dan Pendidikan Matematika Vol 6, No 2 (2021): September 2021 - Februari 2022
Publisher : Universitas Pesantren Tinggi Darul Ulum Jombang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26594/jmpm.v6i2.2228

Persegi ajaib adalah kotak-kotak persegi berisi bilangan berbeda yang disusun sedemikian rupa sehingga jumlah bilangan-bilangan pada baris, kolom, dan diagonal adalah sama. Penelitian ini membahas tentang pola dan algoritma untuk menyusun persegi ajaib berukuran m x m dari rangkaian m^2 bilangan bulat berurutan. Konstruksi algoritma dibagi menjadi tiga kasus, yaitu: algoritma persegi ajaib ordo ganjil (2j + 1) x (2j + 1), algoritma persegi ajaib ordo genap (4j) x (4j), dan algoritma persegi ajaib ordo genap (4j + 2) x (4j + 2) dengan j = 1, 2, ..., m.

Title

Found 30 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 30 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 30 Documents
Search