DEDania Eridani

Dania Eridani

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Dania Eridani has a total of 29 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Philosophical Framework: Integrating Ontology, Epistemology, and Axiology in Edge-AI Flood Early Warning System Design

    by Dania Eridani, Mochamad Agung Wibowo

    The development of the Flood Early Warning System (FEWS) currently uses Artificial Intelligence (AI) to predict real-time floods at the local node with limited resources. However, the literature does not yet explore ontology, epistemology, and axiology in FEWS. This manuscript provides a comprehensive philosophical framework that integrates those three dimensions to strengthen the research design, methodological coherence, and ethical accountability in Edge-AI-based FEWS (EAI-FEWS). Using a conceptual review approach, this manuscript analyzes how ontology commitment in the distributed edge architecture constrains the epistemological method for generating knowledge. How axiology prioritizes equity and response time, how building the system boundary shapes the whole process, and how other dimensions interact across the research process. The framework proposed provides guidance for researchers and practitioners in developing a disaster management system, emphasizing that technical advancement must be grounded in a clear philosophy to produce a rigorous, socially responsible system, with operational viability. This manuscript contributes to the responsible use of AI in critical infrastructure by explaining the philosophical assumptions underlying edge computation in disaster management.

  • Open Access

    A Dynamic-Bayesian-Network-Based Approach to Predict Immediate Future Action of an Intelligent Agent

    by Rinta Kridalukmana, Dania Eridani, Risma Septiana

    Predicting immediate future actions taken by an intelligent agent is considered an essential problem inhuman-autonomy teaming (HAT) in many fields, such as industries and transportation, particularly toimprove human comprehension of the agent as their non-human counterpart. Moreover, the results of suchpredictions can shorten the human response time to gain control back from their non-human counterpartwhen it is required. An example case of HAT that can be benefitted from the action predictor is partiallyautomated driving with the autopilot agent as the intelligent agent. Hence, this research aims to develop anapproach to predict the immediate future actions of an intelligent agent with partially automated drivingas the experimental case. The proposed approach relies on a machine learning method called naive Bayesto develop an action classifier, and the Dynamic Bayesian Network (DBN) as the action predictor. Theautonomous driving simulation software called Carla is used for the simulation. The results show that theproposed approach is applicable to predict an intelligent agent’s three-second time-window immediate futureaction.

  • Open Access

    Sistem Pendukung Keputusan Penentuan Lokasi Embung dengan Menggunakan Metode Višekriterijumsko Kompromisno Rangiranje (VIKOR)

    by Akhmad Ali Sajidin, Ike Pertiwi Windasari, Dania Eridani

    Drought is one of the environmental problems that will specifically cause serious problems if it occurs for a prolonged period. One way to overcome this problem is by building reservoirs or small dam as water storage facilities. In Semarang Regency, there are 8 identified locations for small dam candidates, but not all of them can be built within the 5-year budget plan. Decision support systems can be used to determine the priority of candidate small dams based on predetermined criteria, so that budget resources can be utilized optimally. VIKOR (Višekriterijumsko Kompromisno Rangiranje) is one of the Multi-Criteria Decision Making (MCDM) methods used to solve discrete decision-making problems with conflicting criteria. The VIKOR method can be used to rank candidate small dams with multiple criteria. By implementing the VIKOR method into a decision support system, the process of determining the priority of the best small dams in Semarang Regency can be done effectively and efficiently. The information system was developed using the RAD (Rapid Application Development) method with the PHP programming language and MySQL database, and the Codeigniter framework. In the decision support system, the priority ranking for the construction of small dams in Semarang Regency is displayed based on the results of the VIKOR method calculation using several criteria, including topographic factors (vegetation in inundation area, volume of material embankment, and land acquisition area), hydrological factors (live storage/effective storage volume), effectiveness factors (reservoir lifetime and water cost), and accessibility factors (access road to dam's site). The ranking result generated by the decision support system using VIKOR can be used as a reference by decision makers or organizations to improve efficiency in the process of determining the construction of small dams by providing alternative compromise solutions for the best dam options.

  • Open Access

    Human Action Recognition (HAR) Classification Using MediaPipe and Long Short-Term Memory (LSTM)

    by Ichsan Arsyi Putra, Oky Dwi Nurhayati, Dania Eridani

    Human Action Recognition is an important research topic in Machine Learning and Computer Vision domains. One of the proposed methods is a combination of MediaPipe library and Long Short-Term Memory concerning the testing accuracy and training duration as indicators to evaluate the model performance. This research tried to adapt proposed LSTM models to implement HAR with image features extracted by MediaPipe library. There would be a comparison between LSTM models based on their testing accuracy and training duration. This research was conducted under OSEMN methods (Obtain, Scrub, Explore, Model, and iNterpret). The dataset was preprocessed Weizmann dataset with data preprocessing and data augmentation implementations. Video features extracted by MediaPipe: Pose was used in training and validation processes on neural network models focusing on Long Short-Term Memory layers. The processes were finished by model performance evaluation based on confusion matrices interpretation and calculations of accuracy, error rate, precision, recall, and F1score. This research yielded seven LSTM model variants with the highest testing accuracy at 82% taking 10 minutes and 50 seconds of training duration.


  • Open Access

    Rancang Bangun Sistem Monitoring Dan Controlling Tambak Udang Windu Dengan Konsep Internet Of Things Menggunakan Protokol Message Queuing Telemetry Transport

    by Dania Eridani, Eko Didik Widianto, Nur Kholid

    Udang windu merupakan salah satu jenis udang asli dari Indonesia. Pembudidayaan udang windu sangat dipengaruhi oleh kualitas air pada tambak udang windu. Berdasarkan faktor tersebut, maka dibuatlah sistem yang mampu memantau dan mengontrol kualitas air pada tambak udang windu secara kontinyu dan real-time menggunakan konsep Internet of Things dengan protokol Message Queuing Telemetry Transport. Sistem yang dibangun terdiri dari 2 bagian, yang pertama adalah NodeMCU sebagai primary node yang terhubung dengan sensor (HC-SR04, SEN0161, dan DS18B20) untuk pemantauan kualitas air dan aktuator (motor DC sebagai kincir air). Bagian kedua adalah Raspberry Pi 3 Model B sebagai MQTT broker dan berfungsi untuk mengirimkan hasil pembacaan sensor menuju database. Hasil dari penelitian ini adalah sistem dapat memantau kualitas air dan juga melakukan kontrol terhadap kincir air melalui aplikasi berbasis website. Primary node juga bisa berkomunikasi dengan broker melalui protokol MQTT.


  • Monitoring System in Lora Network Architecture using Smart Gateway in Simple LoRa Protocol

    by Dania Eridani, Eko Didik Widianto, Richard Dwi Olympus Augustinus, Al Arthur Faizal

    LoRa is a low power wide area network communication that has ability to transmit data in a long range with a low power. Lora Alliance has designed a network architecture that usually used, called LoRaWan. The implementation of LoRaWan itself has not meet the requirement of LoRa as one of the LPWAN protocol because it still depend on the internet in processing and displaying data. To meet the purpose of Low Power Wide Area Network (LPWAN) such as simple implementation, low cost, and simple architecture, this research using Simple LoRa Protocol (SLP) with Raspberry Pi 3b+ as the smart gateway. The module used is Dragino LoRa that work in 868 MHz with Spreading Factor 7 and Arduino UNO for the client. The architecture build used in this research implemented to a monitoring system. The system build only to access a Local Area Network. Result showed that the smart gateway could handle the LoRa communication and the monitoring information system together. It also showed that the system average throughput is 489 bit/s and has packet loss 26% at 1-meter distance.


  • Augmented Reality Technology As One Of The Media In Therapy For Children With Special Needs

    by Kurniawan Teguh Martono, Dania Eridani, Adnan Fauzi, Asri Purwanti, Muhammad Ulil

    Learning is a need that must be fulfilled by each human. However, in its development, each human is facing a difference from one to each other. Say, in real life, there are some children with a growth different from the condition in common. The difference in this growth is called with the condition of children with special needs. The existence of children with special needs requires a special treatment. A variety of the methods are used to help children with special needs to growth optimally. The development of computer technology has provided a new breakthrough in helping the children with special needs in improving their motoric ability. One of the breakthroughs used is the development of Medical Rehabilitation Tool by using the augmented reality technology. This technology is able to give a new experience for its user in that the users can directly interact with the virtual objects so as to be able to increase the perception in understanding a form of an object. In this research, the process of testing on the usability of the application to 30 respondents was performed consisting of medical rehabilitation staffs for the speech therapy and occupational therapy. The test on the usability covered the test of Usefulness, Ease of Use, Ease of Learning and Satisfaction. The results of the test from the usefulness aspect showed the result of 81.06%, the aspect of Ease of Use showed the result of 79.13%, the aspect of Ease of Learning showed the result of 78.70 % and the aspect of Satisfaction showed the result of 78.40%. With those results obtained, the respondents agreed that the level of usability from the application can help children with special needs to study.


  • Performance of Sensors Monitoring System using Raspberry Pi through MQTT Protocol

    by Dania Eridani, Eko Didik Widianto

    This paper shows the performance of sensors monitoring system built in Raspberry Pi as implementation of Internet of Things through MQTT protocol. The main focus is to observe the data read by the sensors that showed in the monitoring system over MQTT protocol. The controller used in this research is Raspberry Pi, as it also can directly be used as the transmission board. Temperature sensor and IR Range Sensor used in this research. The result show that the use of QoS parameter in MQTT configuration is the key of the data transfer success. By using the QoS parameter in the MQTT protocol give any packet loss. All of the data from the hardware can be transmitted to the monitoring system. The average delay between the data read by the sensor and the data seen in the HiveMQ websocket client is 0.23s maximum.


  • Data Comparison of NFC PN532 on Wemos Dl and MKR1000 Board through MQTT Protocol

    by Dania Eridani, Eko Didik Widianto, Risma Septiana, Erwan Yudi Indrasto, Kurniawan Teguh Martono, Adnan Fauzi

    This research focused on the comparison of data transfer process in NFC PN532 using two different transmission boards. Wemos Dl and MKR1000 used as the transmission board. The network built with MQTT protocol by using mosquitto message broker. NFC PN532, a transceiver module who works at l3,56MHz and supports several host interfaces used in this paper. SPI used as the interface between transmission board and NFC PN532. The testing mechanism is comparing the data received in server form NFC PN532 that transmitted through both MKR1000 and Wemos D1. The result showed that the use of one topic or two topics in MQTT protocol does not show big difference, but the data transmission using MKR100 is quite better and more balance than Wemos D1.