Search

Reinforcement learning has been implemented to model a task by doing the task repeatedly to get the maximum results based on the reward and punishment policy. It has been implemented in the game and agent-based modelling. In the game, the game agent or Non-Players character can be modelled using several techniques to achieve the goal (e.g. reinforcement learning, deep neural network and Monte Carlo Tree Search). Deep neural networks and Monte Carlo Tree Search, two more sophisticated techniques in reinforcement learning algorithms, assisted the present reinforcement learning in resolving more challenging issues. However, this area has two challenges: the minimum number of data to model and generalization to different environments. Determining a minimum number of data required by the architecture to train the model is quite a cumbersome task to be applied to real- world jobs and situations since it demands substantial data to be explicitly provided and trial-and-error re-configuration. This work proposes a Data-Efficient Reinforcement Learning model by augmenting the data and implementing episodic memory. To illustrate the effectiveness of the proposed model, this research compares it to several models, such as the Deep Q-Network (DQN) with episodic memory to the same model with data augmentation and episodic memory. The model adds to the observations before being stored in the agent's memory, causing the agent to use the same logic and take the same action in comparable situations. The outcome demonstrates that the augmented model can surpass the fundamental model in speed (with an improvement of 50% quicker). [ABSTRACT FROM AUTHOR]

Augmented Reality has become a bridge from the virtual to the real world. This technology has been an implementable choice for fashion companies to provide more information and customization for their user's shopping experience. From the COVID-19 pandemic alongside technological globalization, fashion industries have to find a new way for customers to purchase products. All things considered, AR helped fashion industries keep customers attracted and satisfied. This paper is conducted in a Systematic Literature Review manner to review and compare the AR technologies used by fashion industries. Furthermore, the result would be a proposition of which Augmented Reality technology or platform option is feasible and implementable for the fashion industry. After comparing the technologies used in past papers and their technical and economic feasibility, we propose a 3-term plan. The short-term plan with zero cost and mobile platform-focused application, the medium-term plan with an upgraded version of the garment object from 2D images to 3D objects, and the long-term plan changing the platform used from mobile to a new device. We also conclude that artificial intelligence can also support and improve the augmented reality experience for customers in the fashion retail industry by implementing an AI-powered virtual assistant. [ABSTRACT FROM AUTHOR]

Augmented Reality is a technology that overlays computer-generated information and images in a natural environment to enhance or augment the contextual perception of the user's environment. The utilization of AR is currently one of the most important in developing several industrial sectors, such as architecture, engineering, and construction development. Nowadays, the property sector industry has always been an issue of sale for users that do not have direct access to the property. The pandemic and the economic problem in different countries have also reduced consumer interest in purchasing houses. Due to these issues, this paper aims to provide a platform for addressing these pain areas. The software can apply Augmented Reality to real estate in real-time using a mobile device. Not only can they see the outside of the property, but they can also see the interior design, furniture design, and property mapping in 3D. Our test results show that an application is needed to provide the property sector with high demands. Over 80% of our respondents confirm that our application's presence will help ease the sales problem. Because it can save the customer time, viewing property frequency, and cost saving. These increased demands, we are developing an AR-based application for the market to use and gain more traction in the property sector. [ABSTRACT FROM AUTHOR]

Indonesia is known for its iconic tourism industry. However, there is a significant popularity difference between each destination. This imbalance has been a problem that the Indonesian Ministry of Tourism has been trying to solve through the 'Wonderful Indonesia" campaign. Unfortunately, some did not perceive it as a success. With the breakthrough of Augmented Reality (AR) that is seen as a great factor in enhancing user experience in the tourism sector, this research aims to design and develop an immersive android-based application that implements Augmented Reality and Audio Guidance to promote and create branding for Indonesian tourism spots. This research uses a methodology divided into 4 phases; Analysis Requirement, Research and Collect Data, System Design and Development, and Testing. A survey was conducted in the Analysis Requirement phase to find people's knowledge and preference for tourism spots in Indonesia. Another survey was conducted in the Testing phase to test the application's usability and whether the aim has been achieved. The main findings indicate that most local tourist respondents stated that the application developed is an effective way for them to explore and educate themselves about Indonesian tourism sites. Therefore, this application might be a sustainable answer and a new approach compared to what the Indonesian government has done in the past years. [ABSTRACT FROM AUTHOR]

Recognising emotions from a conversation is a paramount task for a machine to understand the full context of the conversation. However, recognising emotions is not an easy task for a machine that cannot understand social context. This research aims to explore several conventional machine learning methods in emotion recognition modelling, such as Na¨ıve Bayes, K-Nearest Neigh- bour, Decision Trees and Random Forest. Moreover, this research also presents the exploration of deep learning architecture such as Vanilla Deep Neural Networks, Convolutional Neural Networks and Long-Short Term Memory. Moreover, five Mel-Frequency Cepstral Coefficients numbers are also explored (32, 40, 48, 64, 128) with two sampling techniques (Fast Fourier Transform and Kaiser Fast). The results demonstrate that the best training accuracy is achieved by several algorithms such as KNN, ANN and LSTM (100%). Moreover, the model trained with ANN (73%) achieves the best training accuracy. [ABSTRACT FROM AUTHOR]

After facing the COVID-19 pandemic, national and local governments in Indonesia realized a gap in the distribution of health care and human health practitioners. This research proposes two unsupervised learning methods, K-Means and Fuzzy C-Means (FCM), for clustering a list of hospital data in Jakarta, Indonesia, which contains information about the number of its human health resources. The datasets used in this study were obtained from the website the Ministry of the Health Republic of Indonesia provided through the content scraping method. The result shows that implementing K-Means and FCM clustering results in the same number of clusters. Nevertheless, both results have different areas and proportions that can be observed by three distance metrics, such as Hamming, Euclidean, and Manhattan distance. By using the clustering result using the K-Means algorithm, the hospital list was separated into three clusters with a proportion of 84.82%, 14.66%, and 0.52% for clusters 0, 1, and 2, respectively. Meanwhile, using the FCM algorithm, the hospital list was separated into three clusters with a proportion of 17.80%, 73.82%, and 8.38% for clusters 0, 1, and 2, respectively. To the best of our knowledge, this is the first discussion of clustering healthcare facilities in Indonesia, especially hospitals, based on their health professionals. [ABSTRACT FROM AUTHOR]

In the current time, Virtual Reality (VR) technologies are improving rapidly. Virtual Reality (VR) allows the user to experience immersive environment that they can interact with through the creation of a 3D digital world. Many studies have explored the potential efficacy of using Virtual Reality (VR) as a form of exposure therapy to treat patients with mental health problems. In this paper, we explored the efficacy of using Virtual Reality Exposure Therapy (VRET) with other extra intervention (e.g., medicine, extra hardware, etc.) to treat patients diagnosed with PTSD. The result indicates that some extra intervention does influence the outcome by seeing the difference in CAPS score. The highest CAPS score difference of 17.7 is achieved by 3MDR which combines the use of VRET (Virtual Reality Exposure Therapy), EMDR (Eye Movement Desensitization and Reprocessing) and walking on a treadmill. [ABSTRACT FROM AUTHOR]

Nowadays, there has been much attention on computer vision regarding human-computer interaction, especially facial expression recognition (FER). Many researchers have explored and suggested systems for this field. In this paper, we propose the Deep Learning architecture to improve the performance of models from the previous work. Additionally, we propose the BiLSTM-CNN model, which combines our proposed CNN and BiLSTM model. Besides that, we also compare the model to our CNN and LSTM-CNN models. We conduct the experiments on the CK+ dataset and evaluate the accuracy rate of the built models. Data augmentation is used in the dataset to improve the model's performance and prevent overfitting. The results demonstrate that the BiLSTM-CNN method achieves a state-of-the-art accuracy rate compared to other methods from previous work. The highest accuracy of 99.43% is reached by the BiLSTM-CNN model with data augmentation. [ABSTRACT FROM AUTHOR]

Emotions hold a paramount role in the conversation, as it expresses context to the conversation. Text/word in conversation consists of lexical and contextual meanings. Extracting emotions from text has been an interesting work recent thees years. With the advancement of machine learning techniques and hardware to support the machine learning process, recognising emotions from a text with machine learning provides promising and significant results. This research aims to explore several popular machine learning to recognise emotions from a conversation in social media. The algorithms proposed in this research are ranged from traditional machine learning to deep learning techniques. The dataset used in this paper is provided by AffectiveTweets, with a baseline of F1Score of 0.71 with word N-grams and SentiStrength. The research contributes extensive explorations in a number of machine learning algorithms, resulting in a total of 2302 features sets were explored, where each features sets has 100-1000 features extracted from the text. The results demonstrate Generalised Linear Model provides the best Accuracy score (0.92), Recall (0.902), Precision (0.902), F1 score (0.901) with standard deviation of accuracy of ±1,2%. [ABSTRACT FROM AUTHOR]

Games have been a popular tool to help to improve learners' and facilitators' learning experiences. Games also have been implemented as the alternative learning tools, where some research shown games were improved the learners' performance and enhance the learners' and facilitators' learning experiences. This paper aims to demonstrate the serious game design to help parents to teach their children of ethics and morality. It is paramount to introduce ethics and morality to the children as soon as possible. Children development at an early age is essential to be occupied by the proper introduction to ethics and morality based on their family, culture, and society. This research proposes a serious game design using the Six Facets of Serious Game Design and Adam's Game Design. With the experts' consultation, a serious game with 12 stages was developed and evaluated. The results show that the game proposed provides a fun alternative for their children to learn something while having fun in the game. The respondents also argued that the value introduced in the game was easy to understand and implemented. Most of the respondents were having fun in the game and felt that the game is interesting and satisfying. For the future research direction, more complex scenarios will be introduced in the game. [ABSTRACT FROM AUTHOR]

This research aims to implement Affective Computing as part of game design by capturing, processing, and interpreting the player's emotions to enhance the player's experience in the game. We argue by implementing affective computing in the game will statistically enhance the players game experiences. Two almost identical games designed and developed to prove the claim. One game imbued with the affective computing system by capturing players emotions from their facial expressions as the game input, when the other game only implement a touch screen system for the input. Both games then were evaluated in two groups of 50 respondents in each group. A combination of Game Experiences Questionnaire and Immersiveness Game Questionnaire used to evaluate the game player experiences in this research. The result concludes that most of the game experiences score in the game with FER system implemented was statistically increased compared to the one with-out the FER system implemented, except for Q3 (p = 0.06), and Q9 (p = 0.08). [ABSTRACT FROM AUTHOR]

The concept of e-sport has been popular over these years. The e-sport industry reached more than one billion US Dollar of market share in 2019 and expecting significant growth over the next years. E-sports has become one of the most growth potential projects that provide the innovation and may attract investment in this current era of digital-electronics industry. This also leads to the growing number of people who are interested in craving their career in the e-sport industry (e.g. as professional players). However, a recent study shows only a number of people who are aware and know how to pave their way towards professional players profession. Not many people know what to expect, how to prepare, and what to do if they wanted to become a professional player in e-sport industry. Hence, this research aims to provide education about the e-sports' career path by developing a mobile-based game application. The main objective of this application is to make the character become a professional player. The results of this research indicate that the proposed game helps players to understand more on the e-sport industry. The players also become more motivated to become a professional player in e-sport industry after they played the game. [ABSTRACT FROM AUTHOR]

The growth of brain glioma concentration is spatio-temporally modeled by using reaction-diffusion. The model assumes heterogeneity in one dimensional space based on grey and white matter arrangement. In this study, model is combined with three different types of growth function. The exponential function produces a linear model, while the logistic and Gompertzian function produce a nonlinear model. The models are solved numerically by finite difference method Crank-Nicolson scheme. Tumor attributes including maximum concentration, number of cell, mean radial distance, and growth speed are also obtained to observe the growth pattern of brain glioma. Python-based desktop application is developed to simulate the numerical solutions and display the results quantitatively and visually. Analysis of the simulation results show differences in growth pattern based on the parameters used. Based on simulation and analysis, the maximum concentration of Gompertzian growth reaches a turning point that is much faster (about 2.8 times) than the other two. [ABSTRACT FROM AUTHOR]

Linear regression model can be used to describe the relationship between dependent variable and independent variables. But, if there is a spatial effect in the data, it must use a spatial model. This is because the independent variables that influence dependent variable can be different at each location. Spatial autoregressive (SAR) model is a spatial method that can be used to describe the relationship between dependent variable and independent variables by considering spatial impact. Before SAR model is carried out, it must be tested for spatial effect using Moran's I. This is used to ensure that there are spatial effect in the data. Because if there is no spatial effect, then can be use a global model, that is linear regression model using Ordinary Least Square (OLS). In this paper, OLS and SAR model was used on average expenditure data of Papua. Expenditures must be made by each region in Indonesia due to subsidies from the central government. Papua is a province that has the highest average expenditure compared to other provinces. Therefore, the average expenditure becomes important thing that needs to be considered by the government of Papua. This is because there must be equal distribution of expenditure at each district / city in Papua. The result of this research indicated that there is a spatial effect on the average expenditure data in Papua. Comparison result of OLS model and SAR model showed that SAR model is better than OLS model. Because MAPE, AIC and RMSE of SAR model are smaller than OLS. [ABSTRACT FROM AUTHOR]

Recognising characters from text have been a popular topic in the computer vision area. The application can benefit to many problems in the world. For example: recognising text in documents, classifying the text or scripts of documents, plate recognition, etc. Many researchers have been developed the methods for recognising characters in by using Optical Character Recognition methods. Although text recognition problem using Optical Character Recognition has been more or less solved, most of the Optical Character Recognition problem explored is belong to Latin alphabet texts. Meanwhile, there are several languages have non-Latin scripts as the written text. Recognising a non-Latin script is quite challenging as the contour and shape of the text are relatively different with a Latin script text. This research aims to collect datasets for OCR in Javanese characters. A total of 5880 characters were collected and trained with several methods with Tesseract OCR tools. The models then be implemented to a mobile phone (Android based). The highest accuracy (97,50%) achieved by the model was achieved by combining single boundary box for the whole parts of the character and the separate boundary boxes in main body and sandangan parts. [ABSTRACT FROM AUTHOR]

Places classification is one of the points of discussion in the computer vision and robotics community. Some renowned techniques such as local-invariant feature extractors (e.g. Scale-invariant feature transform SIFT, Speeded Up Robust Features SURF), as well as Visual BoW approach were used in place classification problems. Nowadays, deep learning methods such as Convolutional Neural Networks (CNNs) have the advantages towards computer vision problems including place classification problem. Albeit, there are several renowned datasets existed to help the community to learn the models, there is no publicly exists in places dataset for specifically places in Indonesia. This paper presents methodology to collect data of visual places in Indonesia, learn deep features from the data, and classify visual places in Indonesia. We aims to contribute a large dataset as well as deep learning models of places in Indonesian. There are more than 16K images collected and augmented to build the places (specifically places in Indonesia) dataset. The highest accuracy score achieved by the models is 92%. [ABSTRACT FROM AUTHOR]

Player's Experience in the game has been known to be one of the essential keys for the success of the game. There are several methods exist to enhance the player's experiences in the games. One of the unexplored methods is a dynamic balancing system using Facial Expression Recognition. The player's facial expression is captured in real-time while the player is playing the game, and the dynamic balancing system will automatically adjust the game difficulty based on the player's facial expressions. This research aims to empirically explore the implementation of Facial Expression Recognition for a dynamic balancing system to enhance the player's experiences in the game. Two action games (2D and 3D) were developed and evaluated with 60 respondents in two groups. Both groups played the game twice, one with facial expression recognition system as dynamic balancing activated and one without. The results demonstrate that they are statistically significant differences (i.e. improvement) between the baseline and enhanced games with p < 0.01. [ABSTRACT FROM AUTHOR]

Chatbot program has been empirically proven help to improve the engagement with users. Moreover, the implementation of a chat-bot program in the industry helps the company to reduce their operational costs in engaging with their customers and employees. There are still quite a number of problems existed in order to build a human-like chatbot program. Understanding a natural conversation and replying back to the conversation the interlocutors, keeping the conversation flowing naturally is a cumbersome task for a computer. This research aims to design an emotionally realistic chatbot system to enhance the believability of the chatbot using Artificial Intelligence Markup Language (AIML) and Information State. The results show that there is a statistically significant improvement to the chatbot believability in the system that has emotions variables induced compare to the one without emotions. Moreover, 63,33% of the respondents perceived Aero and Iris as two different individuals. The future work of this research is to deploy and have an exploration of the chatbot system to other cases. [ABSTRACT FROM AUTHOR]

This research contributes to data collection for Indonesian Language conversation corpus or dataset collected from conversation in the movies through 1961 subtitles with 1678320 unique words. The dataset collected thus trained with deep learning algorithm, a dual encoder Recurrent Neural Network (RNN), Long-Short Term Memory (LSTM) with Attention mechanism with the best result was 2.37% unknown words, loss rate was 1.66, and perplexity of 4.96, trained with vocabulary size of 24000. The dataset could be used for many machine (deep) learning for Natural Language Processing Problem in Indonesian Language, while the pre-trained model can be integrated into several system such as agent-based system, be they a chat-bot or an Embodied Conversational Agent (ECA). For future work, more data will be collected not only from the movie conversation but also natural human-human conversation in Indonesian Language. [ABSTRACT FROM AUTHOR]

This paper propose a serious game to teach social critique to adults with a desktop horror myth game, as the moral story of the myth fits to the social subject to teach to the player/learner. Social problems have been an interesting subject to talk, discuss, and debate in all around the world. This due to the wave of change and shift in the population cohort, and influence the behaviour differences within population cohort, causing a plenty of social problems in the world. Research shown that in Indonesia, social problem is one of the highlight in the society nowadays. Hence, this research is aiming to contribute to a design of a serious game teach social critique to adults, and the evaluations of the serious game design for social problems. The results demonstrate the participants rated that they are happy with the game. Moreover, from the questionnaire, the average score for all items is 3,968, and the highest score belongs to Q6 (4,223), while the lowest score belongs to Q4 (3,800). Moreover, 81% of the players got the right messages about social critique to the society. Finally, the short interview shows the participants felt that the experience when playing the game empower them to get right messages about social critique to the society, where 90% agreed they have gained knowledge and awareness regarding to Narcotics Psychotropic and Addictive Substance and culture of littering arise in the Indonesia. [ABSTRACT FROM AUTHOR]

This research proposes a dynamic game balancing by using Facial Expression Recognition to enhance the player’s experience when playing the game. Research has shown that players generally express their emotions when playing the game. This allow us to capture the player’s expression from their face and use it to dynamically adjust the game difficulties. A preliminary study was conducted to capture what kind of game that would be suitable to test dynamic balancing. The game with a dynamic balancing system then was developed using Scrum methodology as the software development methodology. Furthermore, the dynamic balancing in the game was simulated in the computer and evaluated, with some of items from Immersive Experience Questionnaire, with players played two versions of game, one with dynamic balancing activated and the other with-out dynamic balancing activated. The results evidently show that there was some statistically significant enhancement in the game with Facial Expression Recognition activated as a game dynamic game balancing compare to the one with-out Facial Expression Recognition. [ABSTRACT FROM AUTHOR]

The purpose of this research is to build an android based application to make easy to find the nearest restaurants where a user can see restaurant profiles, comments, like, amenities rating, hygiene, and services and as well as a media campaign for the restaurants. The method used is by questionnaires, interviews, observation, and literature study. The result of this research is NYAM app that can search for food based on GPS, categories, and types and this app can facilitate the search for suitable eating place where user can see restaurant profiles, comments, like, and amenities rating, hygiene, and service, as well as restaurant owners, can promote a place to eat in this application. Most of the respondents find out that the application was easy to use (94.10%), and features were easy to understand (89.1%). The majority of the respondents (84.10%) will continue to use the NYAM system because they find the NYAM system was helping them to find the best restaurant to have their meals with a decent price and have good tastes (88.40%). [ABSTRACT FROM AUTHOR]

Computer has been assisting human in almost all aspects of daily life. Despite the computer outsmart human in some tasks, it is a social ignorant tool. It does not understand and capable of doing natural conversation with human. To build a system that can naturally understand and communicate with human, it is essential to train the system with natural conversation data. This paper proposes a dataset which consists of natural dyadic conversation in Indonesian language. Where literature suggested that there are exists inadequate number of conversation dataset in Indonesian Language. There are 3164 words (formal and slang (informal nonstandard words) annotated from the recording of five groups, with the largest number of words belong to Food topic (Group 3, 826 words), and the lowest one belongs to Travelling topic (Group 2, 372 words). The dataset contributes a pre-trained conversation model with deep learning (LSTM). The model is trained in 10000 iterations, 128 batches, and 4 hidden layers, resulted in a perplexity of 2.01. [ABSTRACT FROM AUTHOR]

This paper aims to explore the possibility of implementing location-based game using the smartphone’s GPS to enhance the player’s experience. With the GPS technology, the player can play in a real world as well as in the virtual game world. We postulate this thus will enhance the player’s experience. To test the hypothesis, a horror game genre with location-based features was developed. The game was evaluated to thirty participants. The results demonstrate the game allowed the participants to explore a new place with fun. In addition, the participants claimed that the game allow them to enjoy a new experience in playing horror game. The participants described that playing the game was giving them more tension compare to playing a conventional horror game. Moreover, the participants claimed that they feel more familiar with the place they’ve explored during the game. Finally, some improvements could be done to improve the experience enhancement to the game. First, more variation in the storyline, quests and the enemy. As the game evaluated in this paper is served only as a prototype, the variation of the storyline, quests and the enemy (i.e. the ghost) is limited. [ABSTRACT FROM AUTHOR]

Natural Language Processing (NLP) is still considered a daunting task to solve for us, researcher in this field. Specifically, there is not many research has been done in a local language like Indonesian Language. Nowdays, there are hundreds of systems that require NLP as their main functions. This could be a good opportunity for us to explore this opportunity. This paper contributes models from deep learning training in Indonesian conversation using dual encoder LSTM as well as vector representation models trained with three corpora using Skip-gram method. The results show that the models are able to make a good correlation, synonym from a particular word in the words representation of vector models. In addition, the conversation models resulted in 1.07 of perplexity in the Combined model in the 14000 th steps. [ABSTRACT FROM AUTHOR]