Journal of Dinda : Data Science, Information Technology, and Data Analytics

Hangout Places Recommendation System Using Content-Based Filtering and Cosine Similarity Methods

Abdul Raihan — 2024-08-01

Coffee shops are becoming the new normal for friends and coworkers to hang out. Selecting the ideal location to hang out can be exceedingly difficult. There are too many choices, and it can be difficult to know where to begin. Based on this problem, a web application that responds to the growing need for an easy method of finding local hangouts is named Nongkies. With a focus on social interaction and exploration, this platform uses a recommender system to find cafes, restaurants, and entertainment venues easily. Key features include location-based search, category, and details places. Extensive testing has confirmed the reliability of Nongkies, offering user-friendly and accurate search results. This system is a website app that suggests places to users based on their preferences. This application was developed using the cosine similarity method, which is a systematic approach that uses a similar method based on cosine angles. Content that is less alike gets lower rankings, while more similar content gets the highest rankings in recommendations. Moreover, this app helps users find local hangouts and directions to those locations, especially university students, and the selection of places to socialize has a significant effect on students' learning experiences.

Sentiment Analysis of Handling "Klitih" in Yogyakarta Using Naïve Bayes

Windha Mega P Dhuhita — 2024-08-01

Activities that lead to crimes called "klitih" often occur and disturb the community. The community's response to the handling carried out by the regional government also varied. The public expressed this response using various types of social media, one of which was Twitter. This research analyzes sentiment or responses given by the public by utilizing the social media Twitter to collect data. Data in the form of tweets that have been taken will go through text processing. After that, the text will be weighted using two methods as a comparison, namely tf-idf and count vector. Then the data will be divided into training data and test data to proceed to the classification stage. Classification is carried out using the Naïve Bayes algorithm. To evaluate the results of Naïve Bayes classification, researchers used the Confusion Matrix, by comparing weighting methods and dividing the training data and test data into several different ratios, to find out the scenario that produces the best level of accuracy. The sentiment obtained was dominated by negative sentiment at 75.8%, while positive sentiment was 24.2%. By using existing data, it was found that weighting with count vector had an accuracy rate of 82%. Meanwhile, weighting using TF-IDF obtained an accuracy of 80%.

Development of Palm Oil Production and Sales Monitoring System Based On Android

Chikal Fachdiana — 2024-08-01

Palm oil is one of the most widely used vegetable oils in the world. It is used as a raw material for the economic area and contributes to foreign exchange earnings. The palm oil enterprise performs a critical position in Indonesia's economic development, lowering poverty and creating different businesses supporting the enterprise. This paper aims to assist in improving forecasting, essential factor identification, early caution structures, overall performance monitoring, and decision help for bunches of palm production. in this paper, a machine based totally on system learning is created and applied in order to estimate palm production using models with algorithm decision tree and timeseries.

CNN-LSTM for MFCC-based Speech Recognition on Smart Mirrors for Edge Computing Command

Aji Gautama Putrada — 2024-08-01

Smart mirrors are conventional mirrors that are augmented with embedded system capabilities to provide comfort and sophistication for users, including introducing the speech command function. However, existing research still applies the Google Speech API, which utilizes the cloud and provides sub-optimal processing time. Our research aim is to design speech recognition using Mel-frequency cepstral coefficients (MFCC) and convolutional neural network–long short-term memory (CNN-LSTM) to be applied to smart mirror edge devices for optimum processing time. Our first step was to download a synthetic speech recognition dataset consisting of waveform audio files (WAVs) from Kaggle, which included the utterances “left,” “right,” “yes,” “no,” “on,” and “off. ” We then designed speech recognition by involving Fourier transformation and low-pass filtering. We benchmark MFCC with linear predictive coding (LPC) because both are feature extraction methods on speech datasets. Then, we benchmarked CNN-LSTM with LSTM, simple recurrent neural network (RNN), and gated recurrent unit (GRU). Finally, we designed a smart mirror system complete with GUI and functions. The test results show that CNN-LSTM performs better than the three other methods with accuracy, precision, recall, and an f1-score of 0.92. The speech command with the best precision is "no," with a value of 0.940. Meanwhile, the command with the best recall is "off," with a value of 0.963. On the other hand, the speech command with the worst precision and recall is "other," with a value of 0.839. The contribution of this research is a smart mirror whose speech commands are carried out on the edge device with CNN-LSTM.

Application of Artificial Intelligence in the Design of 2D Escape From Pirates Game with A Star Algorithm Search Method

Rahmat Kurniawan — 2024-08-01

This research is designed to provide a challenging gaming experience by integrating strategy and problem-solving elements. The A* algorithm was chosen due to its efficient ability to find the shortest path in a complex search space. The implementation of this algorithm allows the main character to dynamically avoid obstacles and pirate threats and reach the destination in an optimal way. The test results show that the A* algorithm not only improves game performance but also provides a more realistic and challenging experience for the player. For testing this application, using obstacles and measured based on the value of nodes on the game map. Based on the test results, the A Star algorithm was successfully applied when comparing the computations in the game and manual calculations in the Escape from Pirates game in the test. Thus, this research contributes to the development of artificial intelligence-based games and opens opportunities for further innovation in interactive game design.

Mobile Assistant Application for Street Food Consumers in Bandung

Julius Angger Satrio Wicaksono — 2024-08-01

In the dynamic city of Bandung, the lively street food scene has captured the fascination of tourists, offering a diverse selection of tempting dishes. Nevertheless, a persistent challenge arises from the lack of comprehensive details about these street foods, presenting a hurdle for consumers in making well-informed and health-conscious choices. This predicament underscores the necessity for a solution, leading to the introduction of the Mobile Assistant Application for Street Food Consumers in Bandung. Harnessing cutting-edge computer vision technology, this application seeks to provide a solution by furnishing users with an intuitive and effective tool for accessing in-depth information regarding street foods. The outcomes of thorough experimentation highlight the application's success in precisely identifying a wide array of street foods in Bandung. Users benefit from accurate information on ingredients and nutritional values, empowering them to make informed dietary decisions and elevating the overall street food experience in Bandung. This inventive solution not only addresses the prevailing information gap but also contributes to the well-being of consumers, ushering in a healthier and more enlightened food culture in Bandung at the tip of one's finger.

Sentiment Classification of User Reviews for KAI Access Application Using Naive Bayes Method

Rafi Andi Hidayah — 2024-08-04

KAI Access is a train ticket booking application that offers convenience and various features for its users. However, the app has received a low rating of 2.4 out of 5 stars on the Play Store, indicating user dissatisfaction. This study conducts a quantitative sentiment analysis of the KAI Access application based on user sentiments expressed on Twitter. Using the CRISP-DM method, data were collected from Twitter with the Tweepy tool, amassing around 4,000 tweets from June to August 2023. The data underwent a preprocessing stage to ensure the quality and accuracy of the analysis. This stage involved removing duplicate tweets, eliminating retweets, and filtering out emoticons and other non-text elements. In the modeling stage, the Multinomial Naive Bayes Classifier algorithm was employed, achieving an accuracy rate of 84.6%. The model performed better at identifying negative reviews, with a precision of 0.96, recall of 0.86, and an F1-score of 0.91. In contrast, the identification of positive reviews was less effective, with a precision of 0.41, recall of 0.75, and an F1-score of 0.53. These findings shed light on the low ratings for KAI Access, particularly in the context of user reviews. The results of this study provide further understanding regarding the low rating given to KAI Access, particularly in the context of user reviews. By using this classification system, it is hoped that developers can design more specific improvements to enhance the user experience, especially in handling positive reviews which have the potential for performance improvement.

Comparative Analysis of Linear Regression, Decision Tree and Gradient Boosting for Predicting Stock Price of Bank Rakyat Indonesia

Rahma Dwi Ningsih — 2024-08-05

An investment is the placement of a current amount of funds in the hope of generating a profit in the future. There are several types of investments, including stocks, which are attractive options as they can bring a huge return to investors. However, rapidly fluctuating stock prices are influenced by various factors, such as company performance, interest rates, economic conditions, and government policies. In Indonesia, PT Bank Rakyat Indonesia Tbk (BBRI) had the largest profit among the 10 largest banks by the end of March 2024, with a profit of IDR 13.8 trillion. The higher the bank's return, the greater the investor's interest in purchasing the stock, influencing the stock price. The goal of stock price prediction is to forecast the stock's future price in order to increase investors' potential profits. Various methods, such as Linear Regression, Decision Tree, and Gradient Boosting, have been developed for stock price prediction. Comparative analysis is needed to determine the most accurate method in predicting the stock price of People's Bank of Indonesia, using Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Root Mean Squared Error (RMSE), and R-squared metrics (R²). The analysis showed that Linear Regression achieved the highest accuracy with R² of 0.96, MAE of 65.72 and RMSE of 86.74 compared to the Decision Tree and Gradient Boosting models.

Determining Air Quality Influential Parameters Using Machine Learning Techniques

Evita Fitri — 2024-08-06

Air quality is an important issue in public health and the environment. This research aims to develop an air quality prediction model based on PM₁₀ and PM_2.5 parameters using various regression and machine learning approaches. The dataset used includes air pollutant standard index (ISPU) data from a number of stations in the Jakarta area with an observation period from January to April 2024. The research method includes collecting datasets, reviewing literature and testing several models of machine learning techniques. Furthermore, the handling of outliers was carried out using the numeric outliers node and data normalization to prepare the data before dividing the training and testing data. The models evaluated include Linear Regression, Random Forest Regression, Gradient Boosted Trees, and Multilayer Perceptron (MLP), with validation using 10 times cross-validation. The results showed that the Random Forest Regression and Gradient Boosted Trees models provided good prediction performance for both PM₁₀ and PM_2.5 parameters. Random Forest Regression showed the lowest RMSE value on testing data for PM₁₀ (0.048) and PM_2.5 (0.037), while Gradient Boosted Trees showed the lowest RMSE value on training data for PM_2.5 (0.032). The process of handling outliers and normalizing the data successfully improved the prediction accuracy of the model. Suggestions for future research include the exploration of new models, the addition of meteorological and socio-economic variables, and the application of models in real-time air quality monitoring systems.

Development of Sentiment Analysis System of Simple Pol Application on Google Play Store Using Naive Bayes Classifier Method and BERT Prediction

Muhammad Dhito Maulidan — 2024-08-08

Digitalization in public services raises various sentiments that are very dynamic, one example is the Simpel Pol Health Test application made by PT Cipta Sari Arsonia (CSA). The research objective is to obtain useful information from accurate community review sentiments for service improvement and feedback for service providers and application developers. The method used is Naïve Bayes Classifier with Tf-idf weighting, Multinomial Naïve Bayes with review value indicators and review sentences predicted by the BERT method as a determinant of sentiment value whether positive or negative. Sentiment towards the application shows quite encouraging results, from 3000 data analyzed with 1772 positive reviews and 263 negative reviews with 80% training data and 20% test data, the naïve bayes classification model is able to provide a high level of accuracy, which is 88.7% with a precision of 88.5%, recall of 100% and f1-score of 93.9%. The data showed that most people gave a positive response to this application, with the dominant word being 'easy'. This system was developed using the local-based streamlit framework and proved to be quite reliable in developing systems for data processing and web-based data analysis even though the scraping process is slightly longer than the google colab service. Future research is expected to be able to predict data that is positive or negative with several parameters and several sentiment analysis methods at once and their comparison.

Post-Election Sentiment Analysis 2024 via Twitter (X) Using the Naive Bayes Classifier Algorithm

Yessi Mayasari — 2024-08-10

This study examines sentiment related to the topic of Twitter after the 2024 election, where the topic focused on the 2024 presidential election. Where there are a lot of public opinions and comments after the 2024 presidential election. One of them is the phenomenon when Anies-Muhaimin and Ganjar-Mahfud filed a lawsuit with the Constitutional Court (MK) to appeal over suspicions of fraud over the victory of the elected pair Prabowo-Gibran. By applying the Naïve Bayes Classifier algorithm to analyze public sentiment. Through data crawling, preprocessing, feature extraction, and sentiment classification, the study identified the dominant sentiment and its intensity among social media users. This methodology utilizes quantitative data analysis, using Twitter data linked to specific election-related hashtags. The findings reveal a mix of negative and positive sentiments, reflecting diverse public opinion about election results and related political developments. The accuracy of Naïve Bayes Classifier is highlighted, demonstrating its effectiveness in sentiment classification in the context of social media. This research contributes to understanding public sentiment in the political realm and improving methodological approaches in sentiment analysis using machine learning.

Comparison of Fisher-Yates Shuffle and Linear Congruent Algorithms for Question Randomization

Nugroho Dwi Saputro — 2024-08-13

This research aims to compare Fisher-Yates algorithm and Linear Congruent algorithm in generating random numbers or permutations. The test is conducted using the Chi-Square method to evaluate the quality of randomness generated by both algorithms. The Chi-Square value of the shuffling results is calculated and compared with the critical value of Chi-Square at a significance level of 0.05 with a degree of freedom (df) of 4, which is 9.488. The results show that the Chi-Square value for the Fisher-Yates algorithm is 3.8 and for the Linear Congruent algorithm is 4.3, both of which are below the critical value. This indicates that there is not enough evidence to reject the Null Hypothesis (H₀), implying that the difference in randomness quality between the two algorithms is not statistically significant. Therefore, both algorithms are considered to have equivalent performance. The decision to choose one of the algorithms can be based on other considerations such as complexity and efficiency. Further research is recommended to explore the performance of the algorithms under different conditions.

Comparison of Linear Regression and LSTM (Long Short-Term Memory) in Cryptocurrency Prediction

Marisa Istaltofa — 2024-08-15

Abstract

Cryptocurrency, particularly Bitcoin, has become a major topic in the financial and digital trading sectors due to its ability to facilitate direct transactions without intermediaries and the transparency offered by blockchain technology. However, the high volatility of Bitcoin prices necessitates accurate prediction methods to support better investment decisions. This research aims to compare the accuracy of Linear Regression and Long Short-Term Memory (LSTM) methods in predicting Bitcoin prices using historical data from Yahoo Finance. The research process begins with the collection of historical Bitcoin price data from September 17, 2014, to July 15, 2024, followed by data processing that includes cleaning and splitting the dataset into training and test data. Linear Regression and LSTM models are applied to the training data and tested to evaluate their performance in price prediction. The research findings show that the LSTM model significantly outperforms the Linear Regression model in terms of prediction accuracy. The LSTM model records much lower Mean Squared Error (MSE) and Root Mean Squared Error (RMSE), as well as perfect R² scores on both datasets, demonstrating its high precision in prediction. In contrast, the Linear Regression model shows higher errors and lower explanatory power of data variability. These findings indicate that LSTM is more effective in capturing temporal patterns and Bitcoin price fluctuations, offering better accuracy and potentially being more suitable for future cryptocurrency price analysis, providing better guidance for investors in this highly dynamic market.

Document Similarity Using Term Frequency-Inverse Document Frequency Representation and Cosine Similarity

Adi Widianto — 2024-08-12

Document similarity is a fundamental task in natural language processing and information retrieval, with applications ranging from plagiarism detection to recommendation systems. In this study, we leverage the term frequency-inverse document frequency (TF-IDF) to represent documents in a high-dimensional vector space, capturing their unique content while mitigating the influence of common terms. Subsequently, we employ the cosine similarity metric to measure the similarity between pairs of documents, which assesses the angle between their respective TF-IDF vectors. To evaluate the effectiveness of our approach, we conducted experiments on the Document Similarity Triplets Dataset, a benchmark dataset specifically designed for assessing document similarity techniques. Our experimental results demonstrate a significant performance with an accuracy score of 93.6% using bigram-only representation. However, we observed instances where false predictions occurred due to paired documents having similar terms but differing semantics, revealing a weakness in the TF-IDF approach. To address this limitation, future research could focus on augmenting document representations with semantic features. Incorporating semantic information, such as word embeddings or contextual embeddings, could enhance the model's ability to capture nuanced semantic relationships between documents, thereby improving accuracy in scenarios where term overlap does not adequately signify similarity.

Implementation of Android-Based Flutter Framework and Waterfall Method in Building Marketplace Applications (MariUmroh)

Mardiah Ramadhani — 2024-09-10

The MariUmroh application is an Android-based Umrah and Hajj travel marketplace application designed specifically to bring together sellers and buyers in one digital platform. The MariUmroh marketplace application aims to make it easier for prospective pilgrims and Umrah and Hajj travel companies to interact, transact, and promote their products online. In its business processes, the MariUmroh application uses a B2C (Business to Customer) business model. Where business activities are carried out by travel admins to customers using electronic media directly. The development of the Android-based MariUmroh application is built with the Flutter Framework which uses the Dart programming language. In building the MariUmroh marketplace application, the development system uses the waterfall method, in its creation it begins with needs analysis, system design, program code writing, program testing, program implementation and maintenance. In the MariUmroh application which offers various Umrah and Hajj travel packages to the wider community easily and more efficiently, and sellers easily promote their products and don't have to worry about losing consumers, the features provided by the application are also very useful for pilgrims during their worship at holy land. In building this application, MySQL is used as a database for data storage.

Keywords: Marketplace, Umrah and Hajj travel, flutter framework, Waterlfall