• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.886-887
• /
• 2023

This paper is about the development of kiosks that provide four types of service. Simple UI and educational videos solve the complexity of existing kiosks and provide an intuitive and convenient screen to users. In addition, the AR function, which is a three-dimensional form, shows directions and store representative images. After storing user information in the DB, a learning model is generated using user-based KNN collaborative filtering to provide a recommendation menu. As a result, it is possible to increase user convenience through kiosks using metaverse and AI recommendation services. It is also expected to solve digital alienation of social classes who have difficulty using kiosks.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.119-137
• /
• 2022

Recent recommendation system studies apply various deep learning models to represent user and item interactions better. One of the noteworthy studies is ONCF(Outer product-based Neural Collaborative Filtering) which builds a two-dimensional interaction map via outer product and employs CNN (Convolutional Neural Networks) to learn high-order correlations from the map. However, ONCF has limitations in recommendation performance due to the problems with CNN and the absence of side information. ONCF using CNN has an inductive bias problem that causes poor performances for data with a distribution that does not appear in the training data. This paper proposes to employ a Vision Transformer (ViT) instead of the vanilla CNN used in ONCF. The reason is that ViT showed better results than state-of-the-art CNN in many image classification cases. In addition, we propose a new architecture to reflect side information that ONCF did not consider. Unlike previous studies that reflect side information in a neural network using simple input combination methods, this study uses an independent auxiliary classifier to reflect side information more effectively in the recommender system. ONCF used a single latent vector for user and item, but in this study, a channel is constructed using multiple vectors to enable the model to learn more diverse expressions and to obtain an ensemble effect. The experiments showed our deep learning model improved performance in recommendation compared to ONCF.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.445-454
• /
• 2023

Recently, various fault diagnosis studies are being conducted utilizing data from collaborative robots. Existing studies performing fault diagnosis on collaborative robots use static data collected based on the assumed operation of predefined devices. Therefore, the fault diagnosis model has a limitation of increasing dependency on the learned data patterns. Additionally, there is a limitation in that a diagnosis reflecting the characteristics of collaborative robots operating with multiple joints could not be conducted due to experiments using a single motor. This paper proposes an LSTM diagnostic model that can overcome these two limitations. The proposed method selects representative normal patterns using the correlation analysis of vibration and current data in single-axis and multi-axis work environments, and generates residual patterns through differences from the normal representative patterns. An LSTM model that can perform gear wear diagnosis for each axis is created using the generated residual patterns as inputs. This fault diagnosis model can not only reduce the dependence on the model's learning data patterns through representative patterns for each operation, but also diagnose faults occurring during multi-axis operation. Finally, reflecting both internal and external data characteristics, the fault diagnosis performance was improved, showing a high diagnostic performance of 98.57%.

• Science of Emotion and Sensibility
• /
• v.26 no.4
• /
• pp.115-124
• /
• 2023

This study explores the intersection of literary studies and artificial intelligence (AI), focusing on the common theme of human emotions to foster complementary advancements in both fields. By adopting a comparative perspective, the paper investigates emotion as a shared focal point, analyzing various emotion-related concepts from both literary and AI perspectives. Despite the scarcity of research on the fusion of AI and literary studies, this study pioneers an interdisciplinary approach within the humanities, anticipating future developments in AI. It proposes that literary sensibility can contribute to AI by formalizing subjective literary emotions, thereby enhancing AI's understanding of complex human emotions. This paper's methodology involves the terminology-centered extraction of emotions, aiming to blend subjective imagination with objective technology. This fusion is expected to not only deepen AI's comprehension of human complexities but also broaden literary research by rapidly analyzing diverse human data. The study emphasizes the need for a collaborative dialogue between literature and engineering, recognizing each field's limitations while pursuing a convergent enhancement that transcends these boundaries.

• Journal of Korea Game Society
• /
• v.20 no.3
• /
• pp.85-94
• /
• 2020

AI-based smart factory technologies are only increase short-term productivity. To solve this problem, collaborative intelligence combines human teamwork, creativity, AI speed, and accuracy to actively compensate for each other's shortcomings. However, current automation equipmens require high safety measures due to the high disaster intensity in the event of an accident. In this paper, we design and implement a factory safety control system that uses a depth camera to implement workers and facilities in the virtual world and to determine the safety of workers through simulation.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.249-263
• /
• 2023

Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.259-266
• /
• 2023

In this paper, we propose an algorithm that can improve the accuracy performance of collaborative filtering using attribute-based opinion mining (ABOM). For the experiment, a total of 1,227 online consumer review data about smartphone apps from domestic smartphone users were used for analysis. After morpheme analysis using the KKMA (Kkokkoma) analyzer and emotional word analysis using KOSAC, attribute extraction is performed using LDA topic modeling, and the topic modeling results for each weighted review are used to add up the ratings of collaborative filtering and the sentiment score. MAE, MAPE, and RMSE, which are statistical model performance evaluations that calculate the average accuracy error, were used. Through experiments, we predicted the accuracy of online customers' app ratings (APP_Score) by combining traditional collaborative filtering among the recommendation algorithms and the attribute-based opinion mining (ABOM) technique, which combines LDA attribute extraction and sentiment analysis. As a result of the analysis, it was found that the prediction accuracy of ratings using attribute-based opinion mining CF was better than that of ratings implementing traditional collaborative filtering.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.11
• /
• pp.1577-1585
• /
• 2022

As the interest in the 4th industrial revolution and metaverse increases, metaverse with multi edge structure is proposed and noted. Metaverse is a structure that can create digital doctor-like system through a large amount of image processing and data transmission in a multi edge system. Since metaverse application requires calculating performance, which can reconstruct 3-D space, edge hardware's insufficient calculating performance has been a problem. To provide streamable AI software in runtime, image processing, and data transmission, which is edge's loads, needs to be lightweight. Also lightweight at the edge leads to power consumption reduction of the entire metaverse application system. In this paper, we propose collaborative edge-cloud image processing with remote image processing method and Region of Interest (ROI) to overcome edge's power performance and build streamable and runtime executable AI software. The proposed structure was implemented using a PC and an embedded board, and the reduction of time, power, and network communications were verified.

• Korean Journal of Artificial Intelligence
• /
• v.11 no.4
• /
• pp.9-13
• /
• 2023

In this study, we explored the potential of integrating interactive AI callbot technology into the medical consultation domain as part of a broader service development initiative. Aimed at enhancing patient satisfaction, the AI callbot was designed to efficiently address queries from hospitals' primary users, especially the elderly and those using phone services. By incorporating an AI-driven callbot into the hospital's customer service center, routine tasks such as appointment modifications and cancellations were efficiently managed by the AI Callbot Agent. On the other hand, tasks requiring more detailed attention or specialization were addressed by Human Agents, ensuring a balanced and collaborative approach. The deep learning model for voice recognition for this study was based on the Transformer model and fine-tuned to fit the medical field using a pre-trained model. Existing recording files were converted into learning data to perform SSL(self-supervised learning) Model was implemented. The ANN (Artificial neural network) neural network model was used to analyze voice signals and interpret them as text, and after actual application, the intent was enriched through reinforcement learning to continuously improve accuracy. In the case of TTS(Text To Speech), the Transformer model was applied to Text Analysis, Acoustic model, and Vocoder, and Google's Natural Language API was applied to recognize intent. As the research progresses, there are challenges to solve, such as interconnection issues between various EMR providers, problems with doctor's time slots, problems with two or more hospital appointments, and problems with patient use. However, there are specialized problems that are easy to make reservations. Implementation of the callbot service in hospitals appears to be applicable immediately.

• Journal of Advanced Technology Convergence
• /
• v.2 no.4
• /
• pp.43-48
• /
• 2023

Globally, there is a collaborative effort to achieve global carbon neutrality in response to climate change. In the case of South Korea, greenhouse gas emissions are rapidly increasing, presenting an urgent situation that requires resolution. In this context, this study developed a thermal energy collection device named a 'steam trap' and created an AI model capable of predicting future electricity usage by collecting energy usage data through steam traps. The average accuracy of electricity usage prediction with this AI model was 96.7%, demonstrating high precision. Consequently, the AI model enables the prediction and management of days with high electricity consumption and identifies which facilities contribute to elevated power usage. Future research aims to optimize energy consumption efficiency through efficient equipment operation using anomaly detection in steam traps and standardizing energy management systems, with the ultimate goal of reducing greenhouse gas emissions.