Analisis Perbandingan Algoritma Decision Tree, kNN, dan Naive Bayes untuk Prediksi Kesuksesan Start-up

Start-ups have a very important role in economic growth, the existence of a start-up can open up many new jobs. However, not all start-ups that are developing can become successful start-ups. This is because start-ups have a high failure rate, data shows that 75% of start-ups fail in their development. Therefore, it is important to classify the successful and failed start-ups, so that later it can be used to see the factors that most influence start-up success, and can also predict the success of a start-up. Among the many classifications in data mining, the Decision Tree, kNN, and Naïve Bayes algorithms are the algorithms that the authors chose to classify the 923 start-up data records that were previously obtained. The test results using cross-validation and T-test show that the Decision Tree Algorithm is the most appropriate algorithm for classifying in this case study. This is evidenced by the accuracy value obtained from the Decision Tree algorithm, which is greater than other algorithms, which is 79.29%, while the kNN algorithm has an accuracy value of 66.69%, and Naive Bayes is 64.21%.

Perbandingan Faktor-Faktor Yang Mempengaruhi Penggunaan Electronic-Know Your Customer (e-Kyc)

There are many technological developments in banks, one of which is online transactions. To get these transactions, an account should be opened using the electronic know you customer (e-KYC) verification system at banks. This research wants to know the differences in the factors that influence behavioral intentions to use e-KYC to open a bank account for SCB (The Siam Commercial Bank) Thailand and Bank Mandiri Indonesia. This is quantitative research using a survey. We have prepared a questionnaire of 160 respondents: 80 for Bank Mandiri and 80 for SCB. The results indicate that the willingness to use electronic identity verification services influence by the availability of technology, the external impact of the network, safety awareness, perception of trust, and perception of security. The perception of security affects the perception of trust, and technical protection, also the transaction procedure does not affect the perception of trust.

Implementasi Deep Learning untuk Entity Matching pada Dataset Obat (Studi Kasus K24 dan Farmaku)

Data processing speed in companies is important to speed up their analysis. Entity matching is a computational process that companies can perform in data processing. In conducting data processing, entity matching plays a role in determining two different data but referring to the same entity. Entity matching problems arise when the dataset used in the comparison is large. The deep learning concept is one of the solutions in dealing with entity matching problems. DeepMatcher is a python package based on a deep learning model architecture that can solve entity matching problems. The purpose of this study was to determine the matching between the two datasets with the application of DeepMatcher in entity matching using drug data from farmaku.com and k24klik.com. The comparison model used is the Hybrid model. Based on the test results, the Hybrid model produces accurate numbers, so that the entity matching used in this study runs well. The best accuracy value of the 10th training with an F1 value of 30.30, a precision value of 17.86, and a recall value of 100.

Peramalan Pelayanan Service Mobil (After-Sale) Menggunakan Backpropagation Neural Network (BPNN)

The automotive industry in Indonesia, primarily cars, is getting more and more varied. Along with increasing the number of vehicles, Brand Holder Sole Agents (ATPM) compete to provide after-sale services (mobile service). However, the company has difficulty knowing the rate of growth in the number of mobile services handled, thus causing losses that impact sources of income. Therefore, we need a standard method in determining the forecasting of the number of car services in the following year. This study implements the Backpropagation Neural Network (BPNN) method in forecasting car service services (after-sale) and Mean Square Error (MSE) for the process of testing the accuracy of the forecasting results formed. The data used in this study is car service data (after-sale) for the last five years. The results show that the best architecture for forecasting after-sales services using BPNN is the 5-10-5-1 architectural model with a learning rate of 0.2 and the learning function of trainlm and MSE of 0.00045581. This proves that the BPNN method can predict mobile service (after-sale) services with good forecasting accuracy values.

Implementasi Algoritma RC4 pada Sistem Pengamanan Dokumen Digital Soal Ujian

The development of science and technology has led to changes in the use of documents in life to become digital data. However, this can cause problems, namely regarding data security and confidentiality. To increase security and confidentiality can be done with cryptographic algorithm RC4. The research method uses the Waterfall method. The result of this research is a website that can secure document files with * doc extension using the RC4 algorithm. The test was carried out using the blackbox test and the CrackStation test for encryption testing. The results of the test show that the website can run well and successfully implements the RC4 algorithm.

Optimasi Keamanan Web Server terhadap Serangan Broken Authentication Menggunakan Teknologi Blockchain

The aspect of the internet that needs to be considered a security is the login system. The login system usually uses a username and password as an authentication method because it is easy to implement. However, data in the form of usernames and passwords are very vulnerable to theft, so it is necessary to increase the security of the login system. The purpose of this research is to investigate the security of the system. Whether the system is good at protecting user data or not, minimizing execution errors from the system and minimizing risk errors on the system so that the login system can be used safely. This research is conducted to test the system security with Burp Suite on the login system that has been built. Testing the security of this system by experimenting with POST data which is secured using blockchain technology makes the data sent in the form of hash blocks safer and more confidential so that the system is safer than before. Blockchain technology has successfully secured usernames and passwords from broken authentication attacks. By using the Burp Suite testing system, login is more specific in conducting security testing.

Klasifikasi Tingkat Kepuasan Mahasiswa Terhadap Pembelajaran Secara Daring Menggunakan Algoritma Naïve Bayes

Since the spread of Covid-19 in Indonesia, in early March 2020, the activities of Educational Institutions have not been disrupted. As conventional learning. Learning at Singaperbangsa University began with regulation from the Ministry of Education and Culture of the Republic of Indonesia, from learning that boldly affects concentration, influences concentration, such as signals, learning atmosphere, and teaching methods, so that factors affect the level of student satisfaction in learning. This study aims to determine the level of student satisfaction with learning who dares to use the Bayes naive algorithm using RapidMiner tools with results obtained with an accuracy rate of 76.92%, class precision of 100.00%, class recall 57.14%, and an AUC value of 0.881 or close to, so the resulting model is good. In other words, the results obtained using the Naïve Bayes algorithm can be used as material for making decisions about the level of online learning satisfaction.

Perbandingan Algoritma Klasifikasi Sentimen Twitter Terhadap Insiden Kebocoran Data Tokopedia

Public responses, posted on Twitter reacting to the Tokopedia data leak incident, were used as a data set to compare the performance of three different classifiers, trained using supervised learning modeling, to classify sentiment on the text. All tweets were classified into either positive, negative, or neutral classes. This study compares the performance of Random Forest, Support-Vector Machine, and Logistic Regression classifier. Data was scraped automatically and used to evaluate several models; the SVM-based model has the highest f1-score 0.503583. SVM is the best performing classifier.

Segmentasi Pelanggan Berdasarkan Perilaku Penggunaan Kartu Kredit Menggunakan Metode K-Means Clustering

Credit card is one of the payment media owned by banks in conducting transactions. Credit card issuers provide benefits for banks with interest that must be paid. Credit card issuers also provide losses to banks that have agreed to pay not to pay their credit card bills. To request a loan from the bank, a cluster model is needed. This study, proposing a segmentation system in research using credit cards to determine marketing strategies using the K-Means Clustering method and conducting experiments using the 4 methods namely K-Means, Agglomerative Clustering, GMM, and DBSCAN. Clustering is done using 9000 active credit card user data at banks that have 18 characteristic features. The results of cluster quality accuracy obtained by using the K-Means method are 0.207014 with the number of clusters 3. Based on the results obtained by considering 4 of these methods, the best method for this case is K-Means.

Analisis Sentimen Review Halodoc Menggunakan Nai ̈ve Bayes Classifier

Healthcare service has the role to help and serve people to access medical services, i.e. providing medicines, medical consultation, or health control. Healthcare service has been transforming to a digital platform. Halodoc is one of the digital platforms that people can use for free or paid, user can also give reviews of Halodoc’s performance and services on Google Play Store to give feedback that Halodoc can use to evaluate and improve the app. The Google Play Store review is increasing every day. Therefore an analysis for the review with sentiment analysis for Halodoc’s review is needed, first phase of sentiment analysis for the review is preprocessing which has tokenization, transform to lower cases, filter stopword, dan filter token (by length) processes. The data is divided into two positive and negative classes with cross-validation and a k-fold validation value of 10, using Naïve Bayes Classifier algorithm with 81,68% accuracy and AUC 0.756, categorized as fair classification.