• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1759-1768
• /
• 2022

This paper proposes a model to detect depression-related emotions in a user's speech using wellness dialogue scripts provided by AI Hub, topic-specific daily conversation datasets, and chatbot datasets published on Github. There are 18 emotions, including depression and lethargy, in depression-related emotions, and emotion classification tasks are performed using KoBERT and KOELECTRA models that show high performance in language models. For model-specific performance comparisons, we build diverse datasets and compare classification results while adjusting batch sizes and learning rates for models that perform well. Furthermore, a person performs a multi-classification task by selecting all labels whose output values are higher than a specific threshold as the correct answer, in order to reflect feeling multiple emotions at the same time. The model with the best performance derived through this process is called the Depression model, and the model is then used to classify depression-related emotions for user utterances.

• Steel and Composite Structures
• /
• v.45 no.2
• /
• pp.205-218
• /
• 2022

Proper calculation of splitting tensile strength (STS) of concrete has been a crucial task, due to the wide use of concrete in the construction sector. Following many recent studies that have proposed various predictive models for this aim, this study suggests and tests the functionality of three hybrid models in predicting the STS from the characteristics of the mixture components including cement compressive strength, cement tensile strength, curing age, the maximum size of the crushed stone, stone powder content, sand fine modulus, water to binder ratio, and the ratio of sand. A multi-layer perceptron (MLP) neural network incorporates invasive weed optimization (IWO), cuttlefish optimization algorithm (CFOA), and electrostatic discharge algorithm (ESDA) which are among the newest optimization techniques. A dataset from the earlier literature is used for exploring and extrapolating the STS behavior. The results acquired from several accuracy criteria demonstrated a nice learning capability for all three hybrid models viz. IWO-MLP, CFOA-MLP, and ESDA-MLP. Also in the prediction phase, the prediction products were in a promising agreement (above 88%) with experimental results. However, a comparative look revealed the ESDA-MLP as the most accurate predictor. Considering mean absolute percentage error (MAPE) index, the error of ESDA-MLP was 9.05%, while the corresponding value for IWO-MLP and CFOA-MLP was 9.17 and 13.97%, respectively. Since the combination of MLP and ESDA can be an effective tool for optimizing the concrete mixture toward a desirable STS, the last part of this study is dedicated to extracting a predictive formula from this model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.10
• /
• pp.2627-2642
• /
• 2023

Parkinson's disease (PD) is a typical, chronic neurodegenerative disease involving the concentration of dopamine, which can disrupt motor activity and cause different degrees of gait disturbance relevant to PD severity in patients. As current clinical PD diagnosis is a complex, time-consuming, and challenging task that relays on physicians' subjective evaluation of visual observations, gait disturbance has been extensively explored to make automatic detection of PD diagnosis and severity rating and provides auxiliary information for physicians' decisions using gait data from various acquisition devices. Among them, wearable sensors have the advantage of flexibility since they do not limit the wearers' activity sphere in this application scenario. In this paper, an attention-based temporal network (ATN) is designed for the time series structure of gait data (vertical ground reaction force signals) from foot sensor systems, to learn the discriminative differences related to PD severity levels hidden in sequential data. The structure of the proposed method is illuminated by Transformer Network for its success in excavating temporal information, containing three modules: a preprocessing module to map intra-moment features, a feature extractor computing complicated gait characteristic of the whole signal sequence in the temporal dimension, and a classifier for the final decision-making about PD severity assessment. The experiment is conducted on the public dataset PDgait of VGRF signals to verify the proposed model's validity and show promising classification performance compared with several existing methods.

• Smart Media Journal
• /
• v.13 no.6
• /
• pp.44-53
• /
• 2024

Companies in modern society are increasingly recognizing sentiment classification as a crucial task, emphasizing the importance of accurately understanding consumer opinions opinions across various platforms such as social media, product reviews, and customer feedback for competitive success. Extensive research is being conducted on sentiment classification as it helps improve products or services by identifying the diverse opinions and emotions of consumers. In sentiment classification, fine-tuning with large-scale datasets and pre-trained language models is essential for enhancing performance. Recent advancements in artificial intelligence have led to high-performing sentiment classification models, with the ELECTRA model standing out due to its efficient learning methods and minimal computing resource requirements. Therefore, this paper proposes a method to enhance sentiment classification performance through efficient fine-tuning of various datasets using the KoELECTRA model, specifically trained for Korean.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.2
• /
• pp.166-181
• /
• 2003

With the rapid growth of Internet, many devices such as Web TVs, PDAs and Web phones, begin to be directly connected to the Internet. These devices need real-time operating systems (RTOS) to support complex real-time applications running on them. Development of such real-time applications called embedded internet applications, is difficult due to the lack of adequate tools, especially debuggers. In this paper we present a new tracepoint debugging tool for the Qplus-T RTOS embedded system, which facilitates the instrumentations of the real-time software applications with timing trace-points. Compared with traditional breakpoint debugger, this trace-point debugger provides the ability to dynamically collect and record application data for on-line examination and for further off-line analysis. And, the trace-points can also provide the means for assigning new values to the running application's variables, without neither halting its execution nor interfering with its natural execution flow. Our trace-point debugger provides a highly efficient method for adding numerous monitoring trace-points within a real time target application such as Qplus-T internet applications, utilizing these trace-points to monitor and to analyze the application's behavior while it is running. And also, our trace debugger is different from previous one in that we can specify and detect the timing violations using its RTL (Real-Time Logic) trace experiments.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.4
• /
• pp.383-392
• /
• 2000

Currently, multiprocessors are evaluated almost exclusively with scientific applications. Commercial applications are rarely explored because it is difficult to obtain the source codes of commercial DBMS. Even when the source code is available, such as for POSTGRES, understanding the source code enough to perform detailed meaningful performance evaluations is a daunting task for computer architects.To evaluate multiprocessors with commercial applications, we have developed our own DBMS, called EZDB. EZDB is a parallelized DBMS, loosely inspired from POSTGRES, and running on top of a software architecture simulator. It is capable of executing parallel programs written in SQL. Contrary to POSTGRES, EZDB is not intended as a prototype for a production-quality DBMS. Its purpose is to easily run and evaluate the performance of commercial applications on multiprocessor architectures. To illustrate the usefulness of EZDB, we showed the cache performance data collected for the TPC-B benchmark on a shared-memory multiprocessor. The simulation results showed that the data structures exhibited unique sharing characteristics and that their locality properties and working sets were very different from those in scientific applications.

• Journal of Advanced Navigation Technology
• /
• v.14 no.3
• /
• pp.452-458
• /
• 2010

In case of an embedded system with computing resource restrictions such as system power and cpu, the overhead due to displaying data on the computer screen may have a significant influence on the system performance. This paper describes an initialization method for LCD-driving components such as an ARM Core, an LCD controller, and an SPI(serial peripheral interface). It also introduces a pixel display function and a panel display method using virtual screen for reducing the display overhead for an LN2440SBC system with an ARM9-based S3C2440A microprocessor. A virtual screen is a large space of computer memories allocated much larger than those needed for one-time display of an image. Displaying a specific region of a virtual screen is done by assigning it as a view-port region. Such a display is useful in an embedded system when concurrently running tasks produce and display their respective results on the screen; it is especially so when the execution result of each task is partially modified, instead of being totally modified, on its turn and displayed. If the tasks running on such a system divide and make efficient use of the region of the virtual screen, the display overhead can be minimized. For the performance comparison with and without using the virtual screen, two different images are displayed in turn and the amount of time consumed for their display is measured. The result shows that the display time of the former is about 5 times faster than that of the latter.

• Journal of KIISE:Software and Applications
• /
• v.31 no.3
• /
• pp.379-386
• /
• 2004

Video segmentation is a fundamental task in video indexing and it includes two kinds of shot change detections such as the abrupt transition and the gradual transition. The abrupt shot boundaries are detected by computing the image-based distance between adjacent frames and comparing this distance with a pre-determined threshold value. However, the gradual shot boundaries are difficult to detect with this approach. To overcome this difficulty, we propose the method that detects gradual transition in the MPEG compressed video using the HMM (Hidden Markov Model). We take two different HMMs such as a discrete HMM and a continuous HMM with a Gaussian mixture model. As image features for HMM's observations, we use two distinct features such as the difference of histogram of DC images between two adjacent frames and the difference of each individual macroblock's deviations at the corresponding macroblock's between two adjacent frames, where deviation means an arithmetic difference of each macroblock's DC value from the mean of DC values in the given frame. Furthermore, we obtain the DC sequences of P and B frame by the first order approximation for a fast and effective computation. Experiment results show that we obtain the best detection and classification performance of gradual transitions when a continuous HMM with one Gaussian model is taken and two image features are used together.

• The KIPS Transactions:PartD
• /
• v.10D no.3
• /
• pp.447-458
• /
• 2003

The basic assumption of conventional data mining methodology is that the data set of a knowledge discovery process should be fixed and available before the process can proceed. Consequently, this assumption is valid only when the static knowledge embedded in a specific data set is the target of data mining. In addition, a conventional data mining method requires considerable computing time to produce the result of mining from a large data set. Due to these reasons, it is almost impossible to apply the mining method to a realtime analysis task in a data stream where a new transaction is continuously generated and the up-to-dated result of data mining including the newly generated transaction is needed as quickly as possible. In this paper, a new mining concept, open data mining in a data stream, is proposed for this purpose. In open data mining, whenever each transaction is newly generated, the updated mining result of whole transactions including the newly generated transactions is obtained instantly. In order to implement this mechanism efficiently, it is necessary to incorporate the delayed-insertion of newly identified information in recent transactions as well as the pruning of insignificant information in the mining result of past transactions. The proposed algorithm is analyzed through a series of experiments in order to identify the various characteristics of the proposed algorithm.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.3
• /
• pp.364-372
• /
• 2000

Subcategorization, defining dependency relation between predicates and their complements, is an important source of knowledge for resolving syntactic and semantic ambiguities arising in analyzing sentences. This paper describes a Korean subcategorization dictionary, particularly annotated with semantic roles of complements coupled with thesaural semantic hierarchy as well as syntactic dependencies. For annotating roles, we defined 25 semantic roles associated with surface case markers that can be used to derive semantic structures directly from syntactic ones. In addition, we used more than 120,000 entries of thesaurus to specify concept markers of noun complements, and also used 47 and 17 predicate patterns for verbs and adjectives, respectively, to express dependency relation between predicates and their complements. Using a full-fledged thesaurus for specifying concept markers makes it possible to build an effective selectional restriction mechanism coupled with the subcategorization dictionary, and using the standard predicate patterns for specifying dependency relations makes it possible to avoid inconsistency in the results and to reduce the costs for constructing the dictionary. On the bases of these, we built a Korean subcategorization dictionary for frequently used 13,000 predicates found in corpora with the aid of a tool specially designed to support this task. An experimental result shows that this dictionary can provide 72.7% of predicates in corpora with appropriate subcategorization information.