• Journal of Animal Science and Technology
• /
• v.63 no.2
• /
• pp.453-460
• /
• 2021

Oxytocin (OXT) and serotonin (5-HT) are essential neurotransmitters associated with the behavior of animals. Recently, we found that the plasma concentration of OXT is positively correlated with horse docility and friendliness toward humans. However, the relationships between the neurotransmitters and other temperaments such as fearfulness, dominance, and trainability are unknown. This study aimed to identify whether the plasma concentration of OXT or 5-HT is correlated with fearfulness, dominance, and trainability of horses. Blood samples of 34 horses were collected at the Horse Industry Complex Center of Jeonju Kijeon College. The concentration of OXT and 5-HT was measured in the plasma samples using enzyme-linked immunosorbent assays. The fearfulness, dominance, and trainability of horses were scored by three professors who were very familiar with the horses. One-way analysis of variance with the least significant difference post-hoc analysis was used to compare the scores for fearfulness and dominance among groups. The trainability of horses was compared using the student t-test. The 5-HT was negatively correlated with dominance, but it had no relation with fearfulness. The OXT appeared to be negatively correlated with fearfulness and dominance in horses. Furthermore, OXT was positively correlated with the trainability of horses. Additionally, 5-HT appeared to enhance trainability. In conclusion, the concentration of OXT or 5-HT in horse blood plasma can be used as a biomarker to monitor the fearfulness, dominance, or trainability of horses.

• Journal of Animal Science and Technology
• /
• v.65 no.5
• /
• pp.1094-1104
• /
• 2023

Aggression in horses may cause serious accidents during riding and non-riding activities. Hence, predicting the temperament of horses is essential for selecting suitable horses and ensuring safety during the activity. In certain animals, such as hamsters, plasma melatonin concentrations have been correlated with aggressive behavior. However, whether this relationship applies to horses remains unclear. To address this research gap, this study aimed to evaluate differences in the plasma melatonin concentrations among horses of different breeds, ages, and sexes and examine the correlation between plasma melatonin concentrations and the temperament of the horses, including docility, affinity, dominance, and trainability. Blood samples from 32 horses were collected from the Horse Industry Complex Center of Jeonju Kijeon College. The docility, affinity, dominance, and trainability of the horses were assessed by three professional trainers who were well-acquainted with the horses. Plasma melatonin concentrations were measured using an enzyme-linked immunosorbent assay. The consequent values were compared between the horses of different breeds, ages, and sexes using a three-way analysis of variance and least significant difference post hoc test. Linear regression analysis was employed to identify the relationship between plasma melatonin concentrations and docility, affinity, dominance, and trainability. The results showed that the plasma melatonin concentrations significantly differed with breeds in Thoroughbred and cold-blooded horses. However, there were no differences in the plasma melatonin concentrations between the horse ages and sexes. Furthermore, plasma melatonin concentrations did not exhibit a significant correlation with the ranking of docility, affinity, dominance, and trainability.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.1
• /
• pp.23-27
• /
• 2003

Recently attempts to we the syllable as the recognition unit to enhance performance in continuous speech recognition hate been reported. However, syllables are worse in their trainability than phones and the former have a disadvantage in that contort-dependent modeling is difficult across the syllable boundary since the number of models is much larger for syllables than for phones. In this paper, we propose a method to enhance the trainability for the syllables in Korean and phoneme-context dependent syllable modeling across the syllable boundary. An experiment in which the proposed method is applied to word recognition shows average 46.23% error reduction in comparison with the common syllable modeling. The right phone dependent syllable model showed 16.7% error reduction compared with a triphone model.

• Nuclear Engineering and Technology
• /
• v.54 no.4
• /
• pp.1230-1244
• /
• 2022

This paper proposes an algorithm for signal validation using unsupervised methods in emergency situations at nuclear power plants (NPPs) when signals are rapidly changing. The algorithm aims to determine the stuck failures of signals in real time based on a variational auto-encoder (VAE), which employs unsupervised learning, and long short-term memory (LSTM). The application of unsupervised learning enables the algorithm to detect a wide range of stuck failures, even those that are not trained. First, this paper discusses the potential failure modes of signals in NPPs and reviews previous studies conducted on signal validation. Then, an algorithm for detecting signal failures is proposed by applying LSTM and VAE. To overcome the typical problems of unsupervised learning processes, such as trainability and performance issues, several optimizations are carried out to select the inputs, determine the hyper-parameters of the network, and establish the thresholds to identify signal failures. Finally, the proposed algorithm is validated and demonstrated using a compact nuclear simulator.

• Journal of Microbiology and Biotechnology
• /
• v.12 no.4
• /
• pp.687-691
• /
• 2002

To enhance the production of flavonoids [baicalin, wogonin-7-Ο-glucuronic acid (GA)], which are secondary metabolites of Scutellaria baicalensis Georgi(G.) plant cells, a multilayer perceptron control system was applied to regulate the substrate feeding in a fed-batch cultivation. The optimal profile for the substrate feeding rate in a fed-batch culture of S. baicalensis G. was determined by simulating a kinetic model using a genetic algorithm. Process variable profiles were then prepared for the construction of a multilayer perceptron controller that included massive parallelism, trainability, and fault tolerance. An error back-propagation algorithm was applied to train the multiplayer perceptron. The experimental results showed that neurocontrol incorporated with a genetic algorithm improved the flavonoid production compared with a simple fuzzy logic control system. Furthermore, the specific production yield and flavonoid productivity also increased.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.5
• /
• pp.44-50
• /
• 1995

This paper describes a large-vocabulary continuous speech recognition system using continuous hidden Markov models for the Korean language. To improve the performance of the system, we study on the selection of speech modeling units, inter-word modeling, search algorithm, and grammars. We used triphones as basic speech modeling units, generalized triphones and function word-dependent phones are used to improve the trainability of speech units and to reduce errors in function words. Silence between words is optionally inserted by using a silence model and a null transition. Word pair grammar and bigram model based oil word classes are used. Also we implement a search algorithm to find N-best candidate sentences. A postprocessor reorders the N-best sentences using word triple grammar, selects the most likely sentence as the final recognition result, and finally corrects trivial errors related with postpositions. In recognition tests using a 3,000-word continuous speech database, the system attained $93.1\%$ word recognition accuracy and $73.8\%$ sentence recognition accuracy using word triple grammar in postprocessing.

• Journal of Veterinary Clinics
• /
• v.35 no.3
• /
• pp.107-110
• /
• 2018

A 5-year-old Thoroughbred race horse was presented to Busan Korea Racing Authority equine hospital with a 3-year history of a slow-growing left rostral maxillary mass. The location and progressive growth of the mass eventually resulted in poor food prehension, quidding and mouth bit placement. The mass was solitary and hard, and covered by normal smooth oral mucosa. Radiographic examination of the maxillae showed a flocculated and mixed radiolucent lesion protruding outward and displacing the 202 and 203 teeth caudally. The 202 tooth was in normal size and the 203 tooth was hypoplastic on radiography. Under general anesthesia, a partial surgical resection of the mass was performed to minimize functional loss and facilitate prompt return to track. After surgery, there was improvement in food intake, mouth bit placement, and cosmetic appearance. Histopathological examination determined the resected maxillay mass to be an ossifying fibroma. However, there was continued growth of remnant mass in the maxilla. Equine ossifying fibroma is a rare condition and primarily affects the rostral mandible, and less commonly, the maxillae. In this case, the lesion was slow-growing, and caused cosmetic and functional impairments, including poor food intake and reduced trainability. Surgical resection was performed, but the effect of treatment was limited due to advanced size / stage of the tumor. Early dental care is suggested for horse owners to prevent belated identification and improve successful treatment of oral disorders like ossifying fibroma.