• Journal of Convergence for Information Technology
• /
• v.10 no.10
• /
• pp.40-46
• /
• 2020

With the development of the Internet, users share their experiences and opinions. Since related keywords are used witho0ut considering information such as the general emotion or genre of an unstructured document such as a movie review, the sensitivity accuracy according to the appropriate emotional situation is impaired. Therefore, we propose a system that predicts emotions based on information such as the genre to which the unstructured document created by users belongs or overall emotions. First, representative keyword related to emotion sets such as Joy, Anger, Fear, and Sadness are extracted from the unstructured document, and the normalized weights of the emotional feature words and information of the unstructured document are trained in a system that combines CNN and LSTM as a training set. Finally, by testing the refined words extracted through movie information, morpheme analyzer and n-gram, emoticons, and emojis, it was shown that the accuracy of emotion prediction using emotions and F-measure were improved. The proposed prediction system can predict sentiment appropriately according to the situation by avoiding the error of judging negative due to the use of sad words in sad movies and scary words in horror movies.

• Science of Emotion and Sensibility
• /
• v.13 no.1
• /
• pp.215-224
• /
• 2010

This paper is designed to investigate the attributes of sound and emotion resided in traditional Eastern thought by looking into acoustic theories such as Sunguarakron (聲有哀樂論) in Akgi(樂記), Sungmuaerakron(聲無哀樂論) of Haegang and Akhakgebum(樂學軌範), Chongiron(天機論) in Choson(朝鮮) dynasty. Six types of emotions, namely sadness, pleasure, happiness, anger, respect, and, affection (哀心, 樂心, 喜心, 怒心, 敬心, 愛心) which is related with sounds was closely reviewed through Akgi(樂記). Also attributes of sounds such as loudness, sharpness, pitch, roughness, fluctuation strength and pleasantness was corresponded with plain & complicated(單複), pitch, good & bad(善惡) slow & fast(舒疾), loud & quiet(猛靜) respectively. In addition to this, this paper is narrowed down that the basic ideas about sound and emotions of Choson(朝鮮) confucian scholar was based on theory of music and rhythm on Akgi(樂記). Furthermore, the relationship between expressed sound and emotions which was revealed in Chongiron(天機論) has been examined. Finally, various applied research and studies will be promoted through this study, because this study will provide foundation which supports sounds and emotions of Eastern.

• Journal of Cadastre & Land InformatiX
• /
• v.49 no.1
• /
• pp.65-84
• /
• 2019

This study attempted to predict citizens' emotions regarding mortgage rates using machine learning algorithms. To accomplish the research purpose, I reviewed the related literature and then set up two research questions. To find the answers to the research questions, I classified emotions according to Akman's classification and then predicted citizens' emotions on mortgage rates using six machine learning algorithms. The results showed that AdaBoost was the best classifier in all evaluation categories. However, the performance level of Naive Bayes was found to be lower than those of other classifiers. Also, this study conducted a ROC analysis to identify which classifier predicts each emotion category well. The results demonstrated that AdaBoost was the best predictor of the residents' emotions on home mortgage rates in all emotion categories. However, in the sadness class, the performance levels of the six algorithms used in this study were much lower than those in the other emotion categories.

• Dementia and Neurocognitive Disorders
• /
• v.23 no.3
• /
• pp.146-160
• /
• 2024

Background and Purpose: The emotions of people at various stages of dementia need to be effectively utilized for prevention, early intervention, and care planning. With technology available for understanding and addressing the emotional needs of people, this study aims to develop speech emotion recognition (SER) technology to classify emotions for people at high risk of dementia. Methods: Speech samples from people at high risk of dementia were categorized into distinct emotions via human auditory assessment, the outcomes of which were annotated for guided deep-learning method. The architecture incorporated convolutional neural network, long short-term memory, attention layers, and Wav2Vec2, a novel feature extractor to develop automated speech-emotion recognition. Results: Twenty-seven kinds of Emotions were found in the speech of the participants. These emotions were grouped into 6 detailed emotions: happiness, interest, sadness, frustration, anger, and neutrality, and further into 3 basic emotions: positive, negative, and neutral. To improve algorithmic performance, multiple learning approaches were applied using different data sources-voice and text-and varying the number of emotions. Ultimately, a 2-stage algorithm-initial text-based classification followed by voice-based analysis-achieved the highest accuracy, reaching 70%. Conclusions: The diverse emotions identified in this study were attributed to the characteristics of the participants and the method of data collection. The speech of people at high risk of dementia to companion robots also explains the relatively low performance of the SER algorithm. Accordingly, this study suggests the systematic and comprehensive construction of a dataset from people with dementia.

• Science of Emotion and Sensibility
• /
• v.10 no.3
• /
• pp.471-480
• /
• 2007

Negative emotions play an important function as to human's existence. In this research, we employed the audio-visual film clips to induce negative emotions and examined the classified responses in the autonomic nervous system(ANS) due to each negative emotion.30 adults(22.6 years $old{\pm}1.24$, 15 males and 15 females) took part in this experiment. Through the preliminary experiment, 2 minutes film's stimuli were selected as the emotion-induced stimuli. During the period when participants were viewing and listening to the selected movie, EDA and ECG were examined as soon as one stimulus was displayed, participants were tested by completing the psychological appraisals of their experienced emotion due to each emotional stimulus. With regard to the result of analyzing the psychological responses, each negative emotion appropriately and effectively induced its target emotion. While concerning the result of analyzing ANS responses, each negative emotion induced its respective activation in ANS. What is more, compared with other types of negative emotional stimuli, the scaring stimulus induced higher activation of the sympathetic nervour system(SNS) as to the indexes in EDh and ECG. This research made segmentation of ANS responses to each negative emotion, which has its significance.

• Science of Emotion and Sensibility
• /
• v.21 no.1
• /
• pp.177-186
• /
• 2018

The mechanism of emotion is complex and influenced by a variety of factors, so that it is crucial to analyze emotion in broad and diversified perspectives. In this study, we classified neutral and negative emotions(sadness, fear, surprise) using arousal evaluation, which is one of the psychological evaluation scales, as well as physiological signals. We have not only revealed the difference between physiological signals coupled to the emotions, but also assessed how accurate these emotions can be classified by our emotional recognizer based on neural network algorithm. A total of 146 participants(mean age $20.1{\pm}4.0$, male 41%) were emotionally stimulated while their physiological signals of the electrocardiogram, blood flow, and dermal activity were recorded. In addition, the participants evaluated their psychological states on the emotional rating scale in response to the emotional stimuli. Heart rate(HR), standard deviation(SDNN), blood flow(BVP), pulse wave transmission time(PTT), skin conduction level(SCL) and skin conduction response(SCR) were calculated before and after the emotional stimulation. As a result, the difference between physiological responses was verified corresponding to the emotions, and the highest emotion classification performance of 86.9% was obtained using the combined analysis of arousal and physiological features. This study suggests that negative emotion can be categorized by psychological and physiological evaluation along with the application of machine learning algorithm, which can contribute to the science and technology of detecting human emotion.

• Journal of the Institute of Convergence Signal Processing
• /
• v.20 no.1
• /
• pp.24-30
• /
• 2019

Modern people argue that mental health care is necessary because of various factors such as unstable income and conflict with others. Recently, equipments capable of measuring electrocardiogram (ECG) in wearable equipment have been widely used. In the case of overseas, it can be seen as a medical assistant [14]. By using such functions, studies are being conducted to distinguish representative emotions (joy, sadness, anger, etc.) with objective values. However, most studies are increasing accuracy by collecting complex bio-signals in a limited environment. Therefore, we examine the factors that have the greatest influence on the change and discrimination of biometric information on each stimulus.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.1
• /
• pp.1-15
• /
• 2019

Internet of Things (IoT) is commonly referred to as a future internet technology to provide advanced services by interconnecting physical and virtual things, collecting and using many data from them. The IoT platform is a server platform with a common architecture to collect and share the data independent of the IoT devices and services. Recently, oneM2M, the global standards initiative for Machine-to-Machine (M2M) communications and the IoT announced the availability of oneM2M Release 2 specifications. Accordingly, this paper presents a new oneM2M-compliant IoT platform called Mobius 2.0 and proposes its application to collect the biosignal data from wearable IoT devices for emotion recognition. Experimental results show that we can collect various biosignal data seamlessly and extract meaningful features from the biosignal data to recognize two emotions of joy and sadness.

• Journal of the Korean Society for Library and Information Science
• /
• v.51 no.4
• /
• pp.121-138
• /
• 2017

As information users seek multimedia with a wide variety of information needs, information environments for multimedia have been developed drastically. More specifically, as seeking multimedia with emotional access points has been popular, the needs for indexing in terms of abstract concepts including emotions have grown. This study aims to analyze the index terms extracted from Getty Image Bank. Five basic emotion terms, which are sadness, love, horror, happiness, anger, were used when collected the indexing terms. A total 22,675 index terms were used for this study. The data are three sets; entire emotion, positive emotion, and negative emotion. For these three data sets, co-word occurrence matrices were created and visualized in weighted network with PNNC clusters. The entire emotion network demonstrates three clusters and 20 sub-clusters. On the other hand, positive emotion network and negative emotion network show 10 clusters, respectively. The results point out three elements for next generation of multimedia retrieval: (1) the analysis on index terms for emotions shown in people on image, (2) the relationship between connotative term and denotative term and possibility for inferring connotative terms from denotative terms using the relationship, and (3) the significance of thesaurus on connotative term in order to expand related terms or synonyms for better access points.

• Science of Emotion and Sensibility
• /
• v.14 no.1
• /
• pp.147-156
• /
• 2011

Basic emotions such as happiness, sadness, anger, fear, and disgust have been widely used to investigate emotion-specific autonomic nervous system activity in many studies. On the contrary, surprise emotion, Suggested also as one of the basic emotions suggested by Ekman et al. (1983), has been least investigated. The purpose of this study was to provide a description of cardiovascular responses on surprise stimulus using electrocardiograph (ECG) and photoplethysmograph (PPG). ECG and PPG were recorded from 76 undergraduate students, as they were exposed to a visuo-acoustic surprise stimulus. Heart rate (HR), standard deviation of R-R interval (SD-RR), root mean square of successive R-R interval difference (RMSSD-RR), respiratory sinus arrhythmia (RSA), finger blood volume pulse amplitude (FBVPA), and finger pulse transit time (FPTT) were calculated before and after the stimulus presentation. Results show significant increase in HR, SD-RR, and RMSSD-RR, decreased FBVPA, and shortened FPTT. Evidence suggests that surprise emotion can be characterized by vasoconstriction and accelerated heart rate, sympathetic activation, and increased heart rate variability, parasympathetic activation. These results can be useful in developing an emotion theory, or profiling surprise-specific physiological responses, as well as establishing the basis for emotion recognition system in human-computer interaction.