scholarly journals Optimization of Genetic Algorithm Performance Using Naïve Bayes for Basis Path Generation

2017 ◽  
pp. 273-282 ◽  
Author(s):  
Achmad Arwan ◽  
Denny Sagita Rusdianto
Keyword(s):  
Genetic Algorithm ◽  
Test Data ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Training Data ◽  
Algorithm Performance ◽  
Accuracy Test ◽  
Path Testing ◽  
Independent Path ◽  
Number Of Iterations

Basis path testing is a method used to identify code defects. The determination of independent paths on basis path testing can be generated by using Genetic Algorithm. However, this method has a weakness. In example, the number of iterations can affect the emersion of basis path. When the iteration is low, it results in the incomplete path occurences.  Conversely, if iteration is plentiful resulting to path occurences, after a certain iteration, unfortunately, the result does not change. This study aims to perform the optimization of Genetic Algorithm performance for independent path determination by determining how many iteration levels match the characteristics of the code. The characteristics of the code used include Node, Edge, VG, NBD, and LOC. Moreover, Naïve Bayes is a method used to predict the exact number of iterations based on 17 selected code data into training data, and 16 data into test data. The result of system accuracy test is able to predict the exact iteration of 93.75% from 16 test data. Time-test results show that the new system was able to complete an independent search path being faster 15% than the old system.

scholarly journals Automation Of Independent Path Searching using Depth First Search

2018 ◽  
Vol 3 (1) ◽  
pp. 104
Author(s):  
Achmad Arwan ◽  
Denny Sagita
Keyword(s):  
Genetic Algorithm ◽  
Test Data ◽  
Search Algorithm ◽  
Depth First Search ◽  
Accuracy Test ◽  
Path Testing ◽  
Independent Path ◽  
Number Of Iterations

In a basis path testing, there are independent paths that must be passed/tested at least once to make sure there are no errors in the code and ensure all pseudocode have implemented on the code. Previously, the independent path was generated using the Genetic Algorithm, but the number of iterations influenced the likelihood of the emergence of the corresponding the independent path. Besides, the pseudocode was also unable to be used directly since it must be implemented first, this makes finding an independent path longer because it has to implement the code. This research aims to find out how to find the independent path directly from pseudocode using a graph and how well the Depth First Search algorithm in finding the independent path. It was chosen because it was able to find the paths from a point to a particular point in a graph. The result of the system accuracy test was able to find the correct independent path as much as 52 from 76 test data, where the result of accuracy is 68.4% on average.

scholarly journals Perbandingan Metode Klasifikasi Berita Hoaks Berbahasa Indonesia Berbasis Pembelajaran Mesin

Repositor ◽  
2020 ◽  
Vol 2 (5) ◽  
pp. 675
Author(s):  
Muhammad Athaillah ◽  
Yufiz Azhar ◽  
Yuda Munarko
Keyword(s):  
Test Data ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Training Data ◽  
Features Selection ◽  
Comparison Of Algorithms ◽  
Bayes Algorithm

AbstrakKlasifiaksi berita hoaks merupakan salah satu aplikasi kategorisasi teks. Berita hoaks harus diklasifikasikan karena berita hoaks dapat mempengaruhi tindakan dan pola pikir pembaca. Dalam proses klasifikasi pada penelitian ini menggunakan beberapa tahapan yaitu praproses, ekstraksi fitur, seleksi fitur dan klasifikasi. Penelitian ini bertujuan membandingkan dua algoritma yaitu algoritma Naïve Bayes dan Multinomial Naïve Bayes, manakah dari kedua algoritma tersebut yang lebih efektif dalam mengklasifikasikan berita hoaks. Data yang digunakan dalam penelitian ini berasal dari www.trunbackhoax.id untuk data berita hoaks sebanyak 100 artikel dan data berita non-hoaks berasal dari kompas.com, detik.com berjumlah 100 artikel. Data latih berjumlah 140 artikel dan data uji berjumlah 60 artikel. Hasil perbandingan algoritma Naïve Bayes memiliki nilai F1-score sebesar 0,93 dan nilai F1-score Multinomial Naïve Bayes sebesar 0,92. Abstarct Classification hoax news is one of text categorizations applications. Hoax news must be classified because the hoax news can influence the reader actions and thinking patterns. Classification process in this reseacrh uses several stages, namely  preprocessing, features extraxtion, features selection and classification. This research to compare Naïve Bayes algorithm and Multinomial Naïve Bayes algorithm, which of the two algorithms is more effective on classifying hoax news. The data from this research  from  turnbackhoax.id as hoax news of 100 articles and non-hoax news from kompas.com, detik.com of 100 articles. Training data 140 articles dan test data 60 articles. The result of the comparison of algorithms  Naïve Bayes has an F1-score value of 0,93 and Naïve Bayes has an F1-score value of  0,92.

scholarly journals Application Of Naïve Bayes to Predict the Potential of Rain in Ternate City

2021 ◽  
Vol 328 ◽  
pp. 04011
Author(s):  
Alwin Ali ◽  
Amal Khairan ◽  
Firman Tempola ◽  
Achmad Fuad
Keyword(s):  
Test Data ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Training Data ◽  
Bayes Method ◽  
Calculation Results ◽  
Bayes Algorithm ◽  
Correct Data ◽  
Rain Data ◽  
Naive Bayes Method

The amount of rainfall that occurs cannot be determined with certainty, but it can be predicted or estimated. In predicting the potential for rain, data mining techniques can be used by classifying data using the naive Bayes method. Naïve Bayes algorithm is a classification method using probability and statistical methods. The purpose of this study is how to implement the naive Bayes method to predict the potential for rain in Ternate City, and be able to calculate the accuracy of the Naive Bayes method from system created. The highest calculation results with new data with a total of 400 training data and 30 test data, obtained 30 correct data with 100% precision, 100% recall and 100% accuracy and the lowest calculation results with new data with a total of 500 training data and 50 test data, obtained 38 correct data and 12 incorrect data with a percentage of precision 61.29%, recall 100% and accuracy 76%.

scholarly journals KLASIFIKASI KARAKTER BERDASARKAN PENILAIAN SIKAP SOSIAL TEMAN SEJAWAT PADA KURIKULUM 2013 MEMANFAATKAN NAIVE BAYES

2018 ◽  
Vol 10 (2) ◽  
pp. 1021
Author(s):  
Imron Rosadi
Keyword(s):  
Success Rate ◽  
Test Data ◽  
Naive Bayes ◽  
Social Attitudes ◽  
Naïve Bayes ◽  
Training Data ◽  
Social Character ◽  
Mutual Cooperation ◽  
Information Classification ◽  
The Social

This  aim  of  this  researh  is  to  classify  an  assessment  sentence  from  students,  then  classified  to  produce  an information. Classification is used to determine the character of   each student so that teachers have no difficulty in  assessing  the  social  character  of  the  student,  based  on  asesment  orientation  On  Kurikulum  2013.  This research  is empasize  to  processing opinion for the  students  to evaluate their friend in  SMA Negeri 1 Ngimbang Lamongan,  and  the  opinion  to  evaluate  each  student  will  classified  into  6  social  attitudes  that  is  Honest, Discipline,  Responsibility,  Pay  Attention  (mutual  cooperation,  tolerance),  well  manered  and  selsf  confidence, this  research  is  divide  into  2  phase  that is the  process  to produce  training data  (dataset) and the process to classify  opinion (test data). Both off the process are to  at extracting the attributes and  object components which commented in every document and to decide  the classification social attitudes classfication for each student.On the system for this character clasification produce accuracy with the sucess rate from the result of the testing clasficataion use Algoritman Naive Bayes success rate of 72%, this approach paper I use of 80%, the succes rate of 72% because the use of the sentence for rating colleague had the rate of variants are quite low that affect the process off making the data that will be used to the system off the clasificaton of characterer

scholarly journals Analisis Sentimen Ulasan Kedai Kopi Menggunakan Metode Naive Bayes dengan Seleksi Fitur Algoritme Genetika

2021 ◽  
Vol 8 (3) ◽  
pp. 609
Author(s):  
Naziha Azhar ◽  
Putra Pandu Adikara ◽  
Sigit Adinugroho
Keyword(s):  
Genetic Algorithm ◽  
Feature Selection ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Training Data ◽  
Bag Of Words ◽  
Bayes Method ◽  
Coffee Shop ◽  
Coffee Shops ◽  
F Measure

<p class="Abstrak">Di era sekarang, kedai kopi tak hanya dikenal sebagai tempat berkumpul dan menyeruput kopi saja, tetapi kedai kopi telah menjadi tempat yang nyaman untuk belajar dan bekerja. Namun, tidak semua kedai kopi memiliki kualitas yang baik sesuai dengan apa yang diharapkan pelanggan. Ulasan tentang kedai kopi dapat membantu pemilik kedai kopi untuk mengetahui bagaimana respons mengenai produk dan pelayanannya. Ulasan tersebut perlu diklasifikasikan menjadi ulasan positif atau negatif sehingga membutuhkan analisis sentimen. Terdapat beberapa tahap pada penelitian ini yaitu <em>pre-processing</em> untuk pemrosesan ulasan, ekstraksi fitur menggunakan <em>Bag of Words</em> dan <em>Lexicon Based Features</em>, serta mengklasifikasikan ulasan menggunakan metode <em>Naïve Bayes</em> dengan Algoritme Genetika sebagai seleksi fitur. Data yang digunakan pada penelitian ini sebanyak 300 data dengan 210 data sebagai data latih dan 90 data sebagai data uji. Hasil evaluasi yang didapatkan dari klasifikasi <em>Naïve Bayes</em> dan seleksi fitur Algoritme Genetika yaitu <em>accuracy</em> sebesar 0,944, <em>precision</em> sebesar 0,945, <em>recall</em> sebesar 0,944, dan <em>f-measure</em> sebesar 0,945 dengan menggunakan parameter Algoritme Genetika terbaik yaitu banyak generasi = 50, banyak populasi = 18, <em>crossover</em> <em>rate</em> = 1, dan <em>mutation</em> <em>rate</em> = 0.</p><p class="Abstrak"> </p><p class="Abstrak"><em><strong>Abstract</strong></em></p><p class="Abstract"><em>In this era, coffee shops are not only known as a place to gather and drink coffee, but also have become a comfortable place to study and work. However, not all coffee shops are in good quality according to what customers expect. Coffee shop reviews can help coffee shop owners to find out the response to their products and services. These reviews need to be classified as positive or negative reviews so that sentiment analysis is needed. There are several steps in this study, which are pre-processing to process reviews, feature extraction using Bag of Words and Lexicon Based Features, also classifying reviews using the Naïve Bayes method with Genetic Algorithm as a feature selection. The data used in this study were 300 data with 210 data as training data and 90 data as test data. Evaluation results obtained from the Naïve Bayes classification and Genetic Algorithm feature selection are 0.944 for accuracy, 0.945 for precision, 0.944 for recall, and 0.945 for f-measure using the best Genetic Algorithm parameters which are many generations = 50, many populations = 18, crossover rate = 1, and mutation rate = 0.</em></p><p class="Abstrak"><em><strong><br /></strong></em></p>

scholarly journals Structure Extension of Tree-Augmented Naive Bayes

Entropy ◽  
2019 ◽  
Vol 21 (8) ◽  
pp. 721 ◽  
Author(s):  
YuGuang Long ◽  
LiMin Wang ◽  
MingHui Sun
Keyword(s):  
Naive Bayes ◽  
Classification Performance ◽  
Naïve Bayes ◽  
Training Data ◽  
Conditional Probability Distribution ◽  
Independence Assumption ◽  
Bayesian Network Classifiers ◽  
Leibler Divergence ◽  
The Difference ◽  
Structure Extension

Due to the simplicity and competitive classification performance of the naive Bayes (NB), researchers have proposed many approaches to improve NB by weakening its attribute independence assumption. Through the theoretical analysis of Kullback–Leibler divergence, the difference between NB and its variations lies in different orders of conditional mutual information represented by these augmenting edges in the tree-shaped network structure. In this paper, we propose to relax the independence assumption by further generalizing tree-augmented naive Bayes (TAN) from 1-dependence Bayesian network classifiers (BNC) to arbitrary k-dependence. Sub-models of TAN that are built to respectively represent specific conditional dependence relationships may “best match” the conditional probability distribution over the training data. Extensive experimental results reveal that the proposed algorithm achieves bias-variance trade-off and substantially better generalization performance than state-of-the-art classifiers such as logistic regression.

scholarly journals Analisis Metode K-Nearest Neighbors (K-NN) Dan Naive Bayes Dalam Memprediksi Kelulusan Mahasiswa

2021 ◽  
Vol 3 (2) ◽  
pp. 107-113
Author(s):  
Kartarina Kartarina ◽  
Ni Ketut Sriwinarti ◽  
Ni luh Putu Juniarti
Keyword(s):  
Naive Bayes ◽  
Naïve Bayes ◽  
Entry And Exit ◽  
Accuracy Rate ◽  
Bayes Method ◽  
K Nearest Neighbors ◽  
Bayes Methods ◽  
Accuracy Test ◽  
Student Graduation ◽  
Selection Of

In this research the author aims to apply the K-NN and Naive Bayes algorithms for predicting student graduation rates at Sekolah Tinggi Pariwisata (STP) Mataram, The comparison of these two methods was carried out because based on several previous studies it was found that K-NN and Naive Bayes are well-known classification methods with a good level of accuracy. But which one has a better accuracy rate than the two algorithms, that's what researchers are trying to do. The output of this application is in the form of information on the prediction of student graduation, whether to graduate on time or not on time. The selection of STP as the research location was carried out because of the imbalance between the entry and exit of students who had completed their studies. Students who enter have a large number, but students who graduate on time according to the provisions are far very small, resulting in accumulation of the high number of students in each period of graduation, so it takes the initial predictions to quickly overcome these problems. Based on the results of designing, implementing, testing, and testing the Student Graduation Prediction Application program using the K-NN and Naive Bayes Methods with the Cross Validation method, the result is an accuracy for the K-NN method of 96.18% and for the Naive Bayes method an accuracy of 91.94% with using the RapideMiner accuracy test. So based on the results of the two tests between the K-NN and Naive Bayes methods which produce the highest accuracy, namely the K-NN method with an accuracy of 96.18%. So it can be concluded that the K-NN method is more feasible to use to predict student graduation

scholarly journals Peringkasan dan Support Vector Machine pada Klasifikasi Dokumen

2017 ◽  
Vol 9 (4) ◽  
pp. 416 ◽  
Author(s):  
Nelly Indriani Widiastuti ◽  
Ednawati Rainarli ◽  
Kania Evita Dewi
Keyword(s):  
Support Vector Machine ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Training Data ◽  
Support Vector ◽  
Good Reputation ◽  
Multiclass Support Vector Machine ◽  
Simple Logistic ◽  
Better Than

Classification is the process of grouping objects that have the same features or characteristics into several classes. The automatic documents classification use words frequency that appears on training data as features. The large number of documents cause the number of words that appears as a feature will increase. Therefore, summaries are chosen to reduce the number of words that used in classification. The classification uses multiclass Support Vector Machine (SVM) method. SVM was considered to have a good reputation in the classification. This research tests the effect of summary as selection features into documents classification. The summaries reduce text into 50%. A result obtained that the summaries did not affect value accuracy of classification of documents that use SVM. But, summaries improve the accuracy of Simple Logistic Classifier. The classification testing shows that the accuracy of Naïve Bayes Multinomial (NBM) better than SVM

scholarly journals COMPARISON OF NAIVE BAYES ALGORITHM AND C.45 ALGORITHM IN CLASSIFICATION OF POOR COMMUNITIES RECEIVING NON CASH FOOD ASSISTANCE IN WANASARI VILLAGE KARAWANG REGENCY

2020 ◽  
Vol 17 (1) ◽  
pp. 37-42
Author(s):  
Yuris Alkhalifi ◽  
Ainun Zumarniansyah ◽  
Rian Ardianto ◽  
Nila Hardi ◽  
Annisa Elfina Augustia
Keyword(s):  
Decision Tree ◽  
Naive Bayes ◽  
Confusion Matrix ◽  
Total Sample ◽  
Naïve Bayes ◽  
Food Assistance ◽  
Training Data ◽  
Naive Bayes Classifier ◽  
Bayes Classifier ◽  
Naïve Bayes Classifier

Non-Cash Food Assistance or Bantuan Pangan Non-Tunai (BPNT) is food assistance from the government given to the Beneficiary Family (KPM) every month through an electronic account mechanism that is used only to buy food at the Electronic Shop Mutual Assistance Joint Business Group Hope Family Program (e-Warong KUBE PKH ) or food traders working with Bank Himbara. In its distribution, BPNT still has problems that occur that are experienced by the village apparatus especially the apparatus of Desa Wanasari on making decisions, which ones are worthy of receiving (poor) and not worthy of receiving (not poor). So one way that helps in making decisions can be done through the concept of data mining. In this study, a comparison of 2 algorithms will be carried out namely Naive Bayes Classifier and Decision Tree C.45. The total sample used is as much as 200 head of household data which will then be divided into 2 parts into validation techniques is 90% training data and 10% test data of the total sample used then the proposed model is made in the RapidMiner application and then evaluated using the Confusion Matrix table to find out the highest level of accuracy from 2 of these methods. The results in this classification indicate that the level of accuracy in the Naive Bayes Classifier method is 98.89% and the accuracy level in the Decision Tree C.45 method is 95.00%. Then the conclusion that in this study the algorithm with the highest level of accuracy is the Naive Bayes Classifier algorithm method with a difference in the accuracy rate of 3.89%.

Sign in / Sign up

Export Citation Format

Share Document