Sriwijaya Journal of Informatics and Applications (SJIA)
Not a member yet
49 research outputs found
Sort by
Comparison of Certainty Factor (CF) and Case Based Reasoning (CBR) to Diagnose Infertility in Women
Get PDFInfertility has now become a terrible and serious problem for women. Limited information about infertility suffered by women makes it difficult for them to predict the disease they are suffering from. Therefore we need an expert system that can predict infertility in women. The methods used in this research are Certainty Factor (CF) and Case Based Reasoning (CBR) methods. Certainty Factor (CF) is one of the techniques used to overcome uncertainty in decision making. Case Based Reasoning (CBR) is a problem solving method by remembering similar events that happened in the past and then using that knowledge or information to solve new problems. Based on the test results using 25 test data, the accuracy of the expert system for diagnosing infertility in women using the Certainty Factor (CF) method is 92%, while the curation of the expert system for diagnosing infertility in women using the Case Based Reasoning (CBR) method is 76%.
Spelling Detection based on P300 Signal with Convolutional Neural Network (CNN) Algorithm
Get PDFBrain Computer Interface (BCI) is a system that connects the human brain with the outside world for people who have motor skills disability problems. One form of utilization is the P300 speller which is used for character recognition or detection by classifying the P300 signal. The Convolutional Neural Network (CNN) method is a deep learning method that can be used to handle signal problems with ID-CNN. At the initial stage the data signal will be transformed and followed by a duplication process using RandomOverSampling because the amount of data in each class is not balanced. The data will be divided into training, validation, and test data. After that, a training with CNN will be conducted and followed by an evaluation to find the best model. The test results from this study are a good-fitting CNN model with an evaluation value consisting of an accuracy of 94.27%, precision of 90.64%, sensitivity / recall of 98.30%, and f-measure of 94.31%. Based on the test, the CNN method can be used and implemented in authentication detection based on the P300 signal
Securing Text File on Audio Files using Least Significant Bit (LSB) and Blowfish
Get PDFAlong with the development of technology, communication can be done in various ways, one of which is digital messages. But often the messages sent do not reach their destination and are obtained by irresponsible parties. This happens because of the lack of security in the file. For this reason, security is needed so that messages cannot be stolen or seen by other parties. There are various ways to secure messages, including Steganography and Cryptography techniques. This study uses a combination of the Least Significant Bit method and the Blowfish algorithm to secure secret messages in audio files. This research will measure encryption and decryption time, analysis of message file size changes after encryption and decryption, and PSNR value of audio files. The result of encryption using blowfish is a change in the size of the message file caused by the size of the message file is less than the block cipher size, so additional bytes are given so that the message size matches the block cipher size. The speed of the encryption and decryption process using the blowfish algorithm results in an average time for encryption of 547.98ms while the average time for decryption is 538.19ms. The longest time for the encryption process is 557.30ms and the fastest is 534.50ms, while the longest time for the decryption process is 548.74ms and the fastest is 531.46ms. Hiding messages in audio files using LSB produces PSNR values above 30dB
Decision Support Systems For Selection Of Pet Cat Using Preference Selection Index (PSI) & Multi-Objective Optimization On The Basis Of Ratio Analysis (MOORA) Methods
Get PDFIn its evolution, cats have many variants that make adopters confused in determining the right choice. In the early stages of the search there are several common ways that adopters use, such as visiting websites on the internet, reading magazines or books, or directly coming to a pet store. The search process requires money, effort, and time. Therefore, in this final project was built a Decision Support System for Selection of Pet Cat using Preference Selection Index (PSI) & Multi-Objective Optimization On The Basis Of Ratio Analysis (MOORA) Methods which is expected to be able to help adopters to improve cost efficiency, energy and issued time. This system is expected to be able to provide recommendations for the type of pet cat according to the criteria and needs of the adopter. The criteria used include adoption costs, health, nature, weight, and treatment time. The basic concept of the two method is to calculate the weight of the criteria which is then multiplied by a normalized matrix and ranking. Based on the results of usability testing that applies the Technology Acceptance Model (TAM) theory by distributing questionnaire to 69 respondents, the results obtained are 0.92 with a VERY STRONG relationship level, so this system can be considered useful for users
CLASSIFICATION METHODS ON SENTIMENT ANALYSIS OF TOURISTS ON AIRLINES IN TWITTER
Get PDFoai:ojs.sjia.ilkom.unsri.ac.id:article/16Sentiment analysis is one of the knowledge to find the opinions of society towards a topic of discussion particular. Text mining is the science that many performed by individuals or companies to improve performance and fix complaints public against the services or brand trademarks that exist in the world of business. One of them is business flight or airline flights. One of them is public complaints against certain airlines posted on twitter. It is certainly going to greatly affect the airline 's own because , media social is one of the means of advertising and trade are extensive. Machine learning methods such as Logistics Regression, Kneighbors Classifier, Support Vector Classifier (SVC), Decision Tree Classifier, Random Forest Classifier, and Gaussian. Several classification methods are used to compare the performance of each method to see the best results
Expert System to Diagnose Disease in Toddlers Using Dempster Shafer Method
Get PDFChildren, especially toddlers at the age of two months to five years old are more susceptible to disease. Limited information about diseases that attack children makes it difficult for parents to predict the disease that will suffer from their children. Therefore we need an expert system that can predict the disease suffered by children, and the method used in this study is the Dempster Shafer method. The Dempster Shafer method can be implemented into an expert system to combine separate symptoms (evidence) in calculating the probability of a disease. Based on the test results using 250 test data, the accuracy of the expert system for diagnosing diseases in children under five years old using Dempster Shafer method is 94%.Keywords : Expert System, Dempster Shafer, Disease in Toddler
Diagnosis Of Respiratory Tract Infections In Toddlers With Expert System Using Variable-Centered Intelligent Rule System And Certainty Factor Method
Get PDFExpert system can help the experts in diagnose the Respiratory TractInfection For Toddlers. This research have a purpose to build anexpert system for Android with Kotlin language using Variable-Centered Intelligent Rule System and Certainty Factor method, alsoget the accuracy of it. System’s input is a yes or no answer from Yes-No Question with user. This research use 164 patient data of toddlersat UPTD Kenten Laut Banyuasin Health Center and variables which issymptoms that occurs in toddlers such as cough, cold, hard to breathe,fever, and the results of a physical examination conducted by theexpert. Based on test result, the system has 95,52% accuracy whendiagnose ISPA case, and 100% accuracy when diagnose Pneumoniacase. So, it can be concluded that Variable-Centered Intelligent RuleSystem and Certainty Factor method can be used to diagnoserespiratory infections in toddlers
Classification of Emotions on Twitter using Emotion Lexicon and Naïve Bayes
Get PDFSocial media is a means of interaction and communication. One of the social media that is often used is Twitter. Twitter allows its users to express many things, one of which is being a personal media to provide various kinds of expressions from its users such as emotions. Users can express their emotions and sentiments through writing on the status of their social media posts. One method to find out the emotion in the sentence is using the Emotion Lexicon. However, the lexicon-based method is not good at classifying data because not every word contains emotion. So, there's a need to combine it with other classification method such as Naive Bayes. Naïve Bayes relies on independent assumptions to obtain a classification through the probability hypothesis that each class has. The results of the classification test with Emotion Lexicon alone have 46% accuracy, 45% precision, 51% recall and 36% f-measure. While the results of the classification test with Emotion Lexicon and Naïve Bayes resulted in an accuracy of 65%, precision of 77%, recall of 55%, and f- measure of 59%
Comparison Of Shift Reduce Parsing and Left Corner Parsing Algorithm in Sentence Structure Ambiguity Checker
Get PDFIndonesian is the official language of the Republic of Indonesia and the language of the Indonesian nation's unity. Although it is often used, there are still errors in the use that are not in accordance with the applicable rules. One type of error is due to ambiguity which can cause misunderstandings in interpreting a word or sentence. Structural ambiguity is a type of ambiguity that occurs when the structure of words in a sentence can be given more than one grammatical structure. Left Corner Parsing and Shift Reduce Parsing are parsing methods used to classify sentence structure ambiguity. This research involves preprocessing, namely case folding, tokenizing and Part Of Speech Tagging. This study uses 90 testing data labeled with facts, 30 ambiguous sentences and 60 unambiguous sentences. Based on the results of checking the ambiguity of the sentence structure, the Shift Reduce Parsing algorithm produces an accuracy of 71%, precision 70.6%, recall 59%, and f-measure 58.2%. Meanwhile, Left Corner Parsing produces an accuracy value of 70%, precision 68.7%, recall 57.5%, and f-measure 55.8%
Determining The Quality and Production of Fresh Vegetables Using Simple Multi - Attributes Rating Technique (SMART) - Fuzzy Tsukamoto
Get PDFVegetables are one of the most important needs in Indonesia. This is due to the increasing need for healthy food to meet daily needs. With the need for vegetables, the quality and production process are still hampered because it is done manually. Therefore created a system that can help someone determine the quality and production of the right vegetables. This system uses the SMART method and fuzzy Tsukamoto with the criteria and variables of vegetables used to get good quality and production. The SMART and fuzzy Tsukamoto method used a dataset of 20 vegetable commodities. In this study, 4 criteria and 3 variables were used, namely height, soil pH, temperature and age of harvest for quality determination. The production uses the variables of demand, supply and production