Through technological developments, many fintech P2P lending have emerged which are competing to offer convenience in transactions and offer fast processes. To determine a P2P lending platform as a place to invest, one must know in advance about the company profile or the application and programs offered as a whole. This of course will take a long time to select a P2P lending platform. If you choose an inappropriate P2P lending platform, it will result in losses. The purpose of this research is to build a Decision Support System (DSS) for choosing a P2P lending platform by implementing the Complex Proportional Assessment (COPRAS) approach in order to get the right decision and not take a long time. The COPRAS approach has the ability to produce the best alternative which is limited to alternative analysis through alternative assumptions by providing utility judgment so that the attributes of each alternative are arranged based on intervals. Based on the results of the case studies conducted, the highest utility score was Danamas Lender with a score of 100, then followed by Alami Funding Sharia with a score of 99.2338, Accelerant with a score of 89.8827 and Amartha Microfinance with a score of 83.4988. In addition, based on the results of black box testing, it shows that the software can run as it should.

There is a lot of indoor air pollution, especially from cigarette smoke, wall paint, air fresheners and gas. With this situation, the room uses Air Box WP6003 air quality detection device by transmitting information about air quality through visualization index. This study aims to improve prediction accuracy with fuzzy time series methods processed through 2 naïve and moving average models using forecast transformers and without transformers. The level of prediction accuracy is calculated through several metrics, namely Mean Absolute Percentage Error (MAPE), Sum of Squares Error (SSE), Mean Square Error (MSE), Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). These results can be calculated between the actual value and the predicted value. The data used is 204584 data from 4 parameters including Temperature, TVOC, HCHO and CO2. The test results with the difference from the forecast transformer and without transformer are comparable. Temperature value obtained using naïve with transformer from RMSE of 0.158866 and naïve without transformer of 0.782397, data using moving average with transformer obtained by 0.147546 and moving average without transformer of 0.772570. This can be explained by the error analysis that was tried, where the error rate continued to increase so that the experimental results continued to be far from the actual number. From the test results it can be concluded that the accuracy of air quality prediction using naïve forecast transformer is pretty accurate.

The use of the internet not only brings benefits but also harms the ease of access to pornographic content. Pornographic content is not solely available on adult websites but is also spread on social media accessed by many children to adults. As a result, problems arise due to pornography addiction, such as rape, sexual orientation deviation, and sexual crimes against children. Uniquely, pornography addiction is hard to detect objectively compared to drug or alcohol addiction. This study aims to develop a pornography addiction detection system in the form of a chatbot-based mobile application. The choice of chatbot is because this system supports interactive automatic communication patterns and guarantees the privacy of its users. This study adopts a psychological measurement technique Pornography Addiction Screening Tool (PAST), as a benchmark in developing application logic. Meanwhile, chatbot application development uses the waterfall method which has been tested by best practices as a software development method. Tests performed are in a developer environment. The system validation mechanism uses the Black-box testing method to observe the execution of the application when it runs the functionality that was designed beforehand. In addition, the usability measurement level application is adopted from the System Usability Scale (SUS) method. From the test results, it was found that the pornography addiction detection system operates normally and can be used as a supporting medium for handling pornography addiction by psychiatrists.

Every year the government always holds procurement of goods and services (tenders) which are informed through the Electronic Procurement Service (LPSE) or the General Procurement Plan Information System (SIRUP). The process of selecting the type of procurement is still manual, namely by selecting the package category so that it is possible for mistakes to occur such as the type of service procurement into the category of goods procurement type or vice versa. Therefore, this research proposes to use the Natural Language Processing (NLP) method that can classify these packages based on existing categories. The method used is Long Short-Term Memory (LSTM) by comparing existing classification methods such as naïve bayes, logistic regression, decision tree, XG Boost, Gradient Boost, Random Forest and Support Vector Machine. The results obtained by the LSTM method have a higher accuracy than other methods, with an accuracy of 90.25%. With a parameter configuration of 100 units in the LSTM layer, epoch 10, batch size 64 and validation step 5

Snapper is one of the favorite fish for consumption because it has a myriad of benefits for the human body. There are many types of snapper, especially snapper which is often found in Indonesian waters. Knowing the types of snapper is important knowledge because snapper has different characteristics, for example there are snapper that can be consumed and there are also types of snapper that can be cultivated. However, the lack of information and similar types of snapper makes it difficult for people to identify the type of snapper. This study aims to implement a Self-Organizing Map (SOM) artificial neural network for classification of snapper species based on color and texture characteristics. In order to provide information about the snapper object to be classified, color and texture feature extraction is used. In color feature extraction, RGB and HSV parameters are used and for texture features, the Gray Level Co-occurrence Matrix (GLCM) approach is applied. Furthermore, the characteristic results obtained will be classified using the Self-Organizing Map (SOM) algorithm which divides the input patterns into certain classes so that the network output is in the form of classes that have similarities to the given input. Based on the results of the accuracy test, the built model is capable of producing an accuracy of 89.89%. Thus, the SOM model built for image classification of snapper species is in the good category.

Fuel Oil (BBM) is a very vital commodity. Fuel has an important role in people's lives. Because of the importance of fuel in people's lives, fuel is one of the basic needs of the community. The policy of increasing the price of fuel has always been a phenomenon in various media which causes pros and cons in society. The policy of increasing fuel prices has a big impact on society, both direct and indirect consumption. This study aims to explore public opinion, whether it shows negative or positive sentiment in the policy of increasing fuel prices. The increase in Pertamax fuel prices has drawn several opinions from citizens on Facebook social media. Sentiment analysis research was conducted to determine the response to Facebook comments on Brilio.Net accounts in 2022 related to the increase in Pertamax fuel prices with a dataset of 799 data, as well as a comparison of the number of positive, negative, and neutral comments. In addition, in this study to be able to determine the level of performance generated by the nave Bayes classifier method in the test. The author uses 80% of the comment dataset to be used as training data and 20% to be used as test data to be used as machine learning and test data. Then the data is classified by the system using orange data mining tools so as to produce a percentage of positive sentiment as much as 19%, negative sentiment as much as 22% and neutral sentiment as much as 59%. testing with the nave Bayes classifier method obtained the highest percentage accuracy rate of 99% from all datasets.

Menstrual disorders are changes that occur in the cycle, the amount of blood, and other changes related to the menstrual cycle. Menstrual disorders are often considered a common problem that can affect daily activities. Menstrual disorders that are ignored on an ongoing basis can cause various reproductive system disorders and are also an early sign of dangerous diseases. Many women ignore the menstrual disorders they experience because they are constrained in consulting a doctor both in terms of time, distance, and cost. To help overcome these obstacles, an expert system can be a solution. This expert system is made using the certainty factor method which consists of 43 symptoms with 7 types of disease. This system produces output in the form of a percentage of the type of possible disease experienced by the user and suggestions based on the symptoms selected by the user. The expert system built is expected to be a medium of information about menstrual disorders and can contribute to providing benefits to the community, especially women, to make an initial diagnosis of menstrual disorders experienced without being constrained in terms of time, distance, or cost

Coronary Heart Disease (CHD) is a condition in which the heart's blood supply is blocked or disrupted by fat in the coronary arteries. This disease is the most significant cause of death in Indonesia. CHD can be detected based on the Heart Rate Variability (HRV) index of the Photophletysmograph (PPG) signal taken from a smartphone's camera. However, the use of PPG from smartphone to detect CHD is still rare in real-world applications. Moreover, studies on CHD detection based on PPG signal are also difficult to be found in the scientific literature. Currently, the Electrocardiogram (ECG) signal still dominates as a signal for detecting CHD. This research fills this research gap by proposing a study on the feature selection of PPG signal to detect CHD. There are three feature selection methods studied in this research, i.e., Analysis of Variance (Anova), Pearson Correlation, and Recursive Feature Elimination (RFE). Furthermore, a classification algorithm, called as K-Nearest Neighbors, has also been chosen to create a machine learning model based on the PPG features. The experimental results show that the Pearson Correlation feature selection method produces better CHD detection performance compared to the other two algorithms (Anova and RFE). CHD detection performance using the Pearson Correlation produces an accuracy of 90.9%, sensitivity of 75%, and specificity of 100%.

Halim Jaya Kisaran shop is one of the trading businesses that sells people's basic needs such as shoes and bags. This store requires forecasting in an effort to process data and generate accurate reports. This is done because the Halim Jaya Kisaran Store to determine the demand for shoes is unpredictable, so it often experiences a lack of shoes to be purchased in order to serve the demand for shoes needed by customers. In designing this system, the author uses the Double Moving Average (DMA) forecasting method for the decision support process in determining the number of shoe stocks to be sold for the next month, using the calculation method. This forecasting application was created using the Visual Basic Net 2010 programming language with Microsoft Access as the database. From making the system an application can be produced that can control optimal and economical demand and with a high level of accuracy and can predict the number of shoe requests at the Halim Jaya Kisaran Store so that the forecast results can help the store to avoid running out of stock of shoe requests at the Halim Jaya Kisaran Store

The marketplace industry in Indonesia is starting to grow rapidly over time. Behind the convenience, some users still feel dissatisfied when making transactions. Problems that are often encountered are product incompatibility, delivery problems, types of payment methods, unsatisfactory service, to difficult return policies. In addition, many considerations such as prices, discounts, and promos make users confused about choosing the right marketplace to make transactions. This study applies the AHP-TOPSIS method as decision support to determine the best marketplace in Indonesia with the criteria used, namely discounts & promos, services, features, product quality, payment methods, and availability of goods. While the alternatives used are Tokopedia, Shopee, Blibli, Lazada, and Bukalapak. The results of the implementation with the AHP-TOPSIS method resulted in Shopee being the best marketplace recommendation with a preference value of 0.9897, followed by Tokopedia with a value of 0.7289, Lazada with a value of 0.4145, Bukalapak with a value of 0.0641 and Blibli with a value of 0.0059. The results of this study are expected to be useful for the community in making decisions on choosing the marketplace that best suits their needs and circumstances for conducting transactions