• The Korean Journal of Applied Statistics
• /
• v.37 no.1
• /
• pp.73-86
• /
• 2024

A radial chart of Excel is very useful graphical method in delivering information for numerical data. However, it is not easy to discriminate or classify many individuals. In this case, after shaping each individual of a radial chart, we need to apply shape analysis. For a radial chart, since landmarks for shaping are formed as many as the number of variables representing the characteristics of the object, we consider a shape that connects them to a line. If the shape becomes complicated due to the large number of variables, it is difficult to easily grasp even if visualized using a radial chart. Principal component analysis (PCA) is performed on variables to create a visually effective shape. The classification table and classification rate are checked by applying the techniques of traditional discriminant analysis, support vector machine (SVM), and artificial neural network (ANN), before and after principal component analysis. In addition, the difference in discrimination between the two coordinates of generalized procrustes analysis (GPA) coordinates and Bookstein coordinates is compared. Bookstein coordinates are obtained by converting the position, rotation, and scale of the shape around the base landmarks, and show higher rate than GPA coordinates for the classification rate.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.57-67
• /
• 2024

This research investigates training methods for large language models to accurately identify sentiments and comprehend information about increasing and decreasing fluctuations in the financial domain. The main goal is to identify suitable datasets that enable these models to effectively understand expressions related to financial increases and decreases. For this purpose, we selected sentences from Wall Street Journal that included relevant financial terms and sentences generated by GPT-3.5-turbo-1106 for post-training. We assessed the impact of these datasets on language model performance using Financial PhraseBank, a benchmark dataset for financial sentiment analysis. Our findings demonstrate that post-training FinBERT, a model specialized in finance, outperformed the similarly post-trained BERT, a general domain model. Moreover, post-training with actual financial news proved to be more effective than using generated sentences, though in scenarios requiring higher generalization, models trained on generated sentences performed better. This suggests that aligning the model's domain with the domain of the area intended for improvement and choosing the right dataset are crucial for enhancing a language model's understanding and sentiment prediction accuracy. These results offer a methodology for optimizing language model performance in financial sentiment analysis tasks and suggest future research directions for more nuanced language understanding and sentiment analysis in finance. This research provides valuable insights not only for the financial sector but also for language model training across various domains.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.6
• /
• pp.250-258
• /
• 2007

Current speech recognition technology s achieved high performance with the development of hardware devices, however it is insufficient for some applications where high reliability is required, such as voice control of powered wheelchairs for disabled persons. For the system which aims to operate powered wheelchairs safely by voice in real environment, we need to consider that non-voice commands such as user s coughing, breathing, and spark-like mechanical noise should be rejected and the wheelchair system need to recognize the speech commands affected by disability, which contains specific pronunciation speed and frequency. In this paper, we propose non-voice rejection method to perform voice/non-voice classification using both YIN based fundamental frequency(F0) extraction and reliability in preprocessing. We adopted a multi-template dictionary and acoustic modeling based speaker adaptation to cope with the pronunciation variation of inarticulately uttered speech. From the recognition tests conducted with the data collected in real environment, proposed YIN based fundamental extraction showed recall-precision rate of 95.1% better than that of 62% by cepstrum based method. Recognition test by a new system applied with multi-template dictionary and MAP adaptation also showed much higher accuracy of 99.5% than that of 78.6% by baseline system.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.8
• /
• pp.383-388
• /
• 2006

This Paper presents a novel approach to single channel microphone speech enhancement in noisy environments. Widely used noise reduction techniques based on the spectral subtraction are generally expressed as a spectral gam depending on the signal-to-noise ratio (SNR). The well-known decision-directed(DD) estimator of Ephraim and Malah efficiently reduces musical noise under the background noise conditions, but generates the delay of the a prioiri SNR because the DD weights the speech spectrum component of the Previous frame in the speech signal. Therefore, the noise suppression gain which is affected by the delay of the a priori SNR, which is estimated by the DD matches the previous frame rather than the current one, so after noise suppression. this degrades the noise reduction performance during speech transient periods. We propose a computationally simple but effective speech enhancement technique based on the sigmoid type function for the weight Parameter of the DD. The proposed approach solves the delay problem about the main parameter, the a priori SNR of the DD while maintaining the benefits of the DD. Performances of the proposed enhancement algorithm are evaluated by ITU-T p.862 Perceptual Evaluation of Speech duality (PESQ). the Mean Opinion Score (MOS) and the speech spectrogram under various noise environments and yields better results compared with the fixed weight parameter of the DD.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.6
• /
• pp.572-577
• /
• 2009

Large corpus-based concatenating Text-to-Speech (TTS) systems can generate natural synthetic speech without additional signal processing. To prune the redundant speech segments in a large speech segment DB, we can utilize a decision-tree based triphone clustering algorithm widely used in speech recognition area. But, the conventional methods have problems in representing the acoustic transitional characteristics of the phones and in applying context questions with hierarchic priority. In this paper, we propose a new clustering algorithm to downsize the speech DB. Firstly, three 13th order MFCC vectors from first, medial, and final frame of a phone are combined into a 39 dimensional vector to represent the transitional characteristics of a phone. And then the hierarchically grouped three question sets are used to construct the triphone trees. For the performance test, we used DTW algorithm to calculate the acoustic similarity between the target triphone and the triphone from the tree search result. Experimental results show that the proposed method can reduce the size of speech DB by 23% and select better phones with higher acoustic similarity. Therefore the proposed method can be applied to make a small sized TTS.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3
• /
• pp.229-238
• /
• 2009

This paper describes a sound engine of Korean traditional instruments, which are the Gayageum and Taepyeongso, by using a TMS320F2812. The Gayageum and Taepyeongso models based on commuted waveguide synthesis (CWS) are required to synthesize each sound. There is an instrument selection button to choose one of instruments in the proposed sound engine, and thus a corresponding sound is produced by the relative model at every certain time. Every synthesized sound sample is transmitted to a DAC (TLV5638) using SPI communication, and it is played through a speaker via an audio interface. The length of the delay line determines a fundamental frequency of a desired sound. In order to determine the length of the delay line, it is needed that the time for synthesizing a sound sample should be checked by using a GPIO. It takes $28.6{\mu}s$ for the Gayageum and $21{\mu}s$ for the Taepyeongso, respectively. It happens that each sound sample is synthesized and transferred to the DAC in an interrupt service routine (ISR) of the proposed sound engine. A timer of the TMS320F2812 has four events for generating interrupts. In this paper, the interrupt is happened by using the period matching event of it, and the ISR is called whenever the interrupt happens, $60{\mu}s$. Compared to original sounds with their spectra, the results are good enough to represent timbres of instruments except 'Mu, Hwang, Tae, Joong' of the Taepyeongso. Moreover, only one sound is produced when playing the Taepyeongso and it takes $21{\mu}s$ for the real-time playing. In the case of the Gayageum, players usually use their two fingers (thumb and middle finger or thumb and index finger), so it takes $57.2{\mu}s$ for the real-time playing.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.4
• /
• pp.136-143
• /
• 2007

This experiment was performed to find out the effects of the cicadas' songs on the psychological responses in adolescents. As a basic coulee, the experiment to set up 'Acceptable & Unacceptable' was performed. As a further coulee, five kinds of the cicadas' songs heard frequently were analyzed and adjectives expressing the feeling to cicadas' songs were factor-analyzed, and psychological responses to auditory sensations were analyzed through regression equations. As a result, the effect of the Cryptotympana atrata's song, the Meimuna opalifera's song and the traffic noise are similar in the degree of disturbing the meditation but they are less disturbing than the white noise. The experiment for adjectives expressing was performed, because it is possible that cicadas' songs affect adolescents as a noise. Cicadas' songs can be expressed with three kinds of factors. First factor is [Annoyance], second factor is [Strength] and third factor is [Rhythm]. The first factor dominates in the songs of the Cryptotympana atrata and the Platypleura kaempferi who generate steady sound, and the third factor dominates in the songs of the Meimuna opalifera, the Leptosemia takanonis and the Oncotympana fuscata who generate fluctuating sounds. The loudness of sound didn't affect on the third factor but the emotional values of the fist and the second factors are linearly proportional to the loudness. The analysis results of the first factor associated with noise showed that the annoyance of adolescents is increased in the order of white noise - the Platypleura kaempferi - the Cryptotympana atrata, if the loudness of sounds are generated equally.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.4
• /
• pp.173-181
• /
• 2007

In this paper, we propose a moving target localization algorithm using acoustic spectrograms. A time-versus-frequency spectrogram provide a information of trajectory of the moving target in underwater. For a source at sufficiently long range from a receiver, broadband striation patterns seen in spectrogram represents the mutual interference between modes which reflected by surface and bottom. The slope of the maximum intensity striation is influenced by waveguide invariant parameter ${\beta}$ and distance between target and sensor. When more than two sensors are applied to measure the moving ship-radited noise, the slope and frequency of the maximum intensity striation are depend on distance between target and receiver. We assumed two sensors to fixed point then form a circle of apollonios which set of all points whose distances from two fixed points are in a constant ratio. In case of three sensors are applied, two circle form an intersection point so coordinates of this point can be estimated as a position of target. To evaluates a performance of the proposed localization algorithm, simulation is performed using acoustic propagation program.

• Science of Emotion and Sensibility
• /
• v.26 no.4
• /
• pp.71-92
• /
• 2023

This study investigates the symbolism and meaning of the bead design, its relationship with the Silla culture, and the conditions of glass bead manufacturing to verify the theory of Silla's production of Silla face-inlaid glass beads with excellent artistry and technology. The research method includes investigating the design analysis, ancient documents, myths, relics, glass, and metal production techniques. Moreover, Hongshan cultural relics and other cases of inlaid glass beads were collected. There are records in the literature that the people of Makhan, Buyeo and Silla of ancient Korea people considered beads as treasures and used them for accessories. It was confirmed that all the design elements of the bead-patterned hair topknot, golden crown, birds, and flower trees were closely related to the myth of Kim Al-ji of Silla, the oviparous tales and the sacred birds and divine beasts of the north. Moreover, the pattern and arrangements were found to be similar in other Silla relics. The origin of beads and face pendants was Hongsan culture, and a stone cast for beads was discovered in Bukpyo of Gojoseon, the lower-level culture of Hajiajeom. In addition, excavating inlaid glass beads from Sik-ri tombs of Korea and a face-inlaid glass bead from Toganmori tombs in Japan confirms the theory of Silla's production. The fact that the Baekje people of ancient Korea had a glass bead manufacturing office in Japan in the fifth centuries suggests that the Silla people also had a manufacturing plant in Java, Indonesia, because this place was a crossroads of Silla's Sea Silk Road and a source of raw materials and labors with a close relationship to Silla. Therefore, the face-inlaid glass bead was indeed self-made by Silla, who possessed the tradition of bead myths and hair topknot, and the high-level skills such as gold crowns and metal inlays.

• Korean Journal of Construction Engineering and Management
• /
• v.25 no.1
• /
• pp.50-61
• /
• 2024

The market for compact houses is growing due to the demand for floor plans prioritizing user needs. However, clients often have difficulty communicating their spatial requirements to professionals including architects because they lack the means to provide evidence, such as spatial configurations or cost estimates. This research aims to create a framework that can translate sketched data-driven spatial requirements into 3D building components in BIM models to facilitate spatial understanding and provide building performance analysis to aid in budgeting in the early design phase. The research process includes developing a process model, implementing, and validating the framework. The process model describes the data flow within the framework and identifies the required functionality. Implementation involves creating systems and user interfaces to integrate various systems. The validation verifies that the framework can automatically convert sketched space requirements into walls, floors, and roofs in a BIM model. The framework can also automatically calculate material and energy costs based on the BIM model. The developed frame enables clients to efficiently create 3D building components based on the sketched data and facilitates users to understand the space and analyze the building performance through the created BIM models.