• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.2
• /
• pp.263-283
• /
• 2024

Users' comments after online shopping are critical to product reputation and business improvement. These comments, sometimes known as e-commerce reviews, influence other customers' purchasing decisions. To confront large amounts of e-commerce reviews, automatic analysis based on machine learning and deep learning draws more and more attention. A core task therein is sentiment analysis. However, the e-commerce reviews exhibit the following characteristics: (1) inconsistency between comment content and the star rating; (2) a large number of unlabeled data, i.e., comments without a star rating, and (3) the data imbalance caused by the sparse negative comments. This paper employs Bidirectional Encoder Representation from Transformers (BERT), one of the best natural language processing models, as the base model. According to the above data characteristics, we propose the F_MixBERT framework, to more effectively use inconsistently low-quality and unlabeled data and resolve the problem of data imbalance. In the framework, the proposed MixBERT incorporates the MixMatch approach into BERT's high-dimensional vectors to train the unlabeled and low-quality data with generated pseudo labels. Meanwhile, data imbalance is resolved by Focal loss, which penalizes the contribution of large-scale data and easily-identifiable data to total loss. Comparative experiments demonstrate that the proposed framework outperforms BERT and MixBERT for sentiment analysis of e-commerce comments.

• Journal of KIISE:Software and Applications
• /
• v.33 no.12
• /
• pp.1090-1102
• /
• 2006

This paper suggests a new algorithm for detecting the plagiarism among a set of source codes, constrained to be functionally equivalent, such are submitted for a programming assignment or for a programming contest problem. The typical algorithms largely exploited up to now are based on Greedy-String Tiling, which seeks for a perfect match of substrings, and analysis of similarity between strings based on the local alignment of the two strings. This paper introduces a new method for detecting the similar interval of the given programs based on an adaptive similarity matrix, each entry of which is the logarithm of the probabilities of the keywords based on the frequencies of them in the given set of programs. We experimented this method using a set of programs submitted for more than 10 real programming contests. According to the experimental results, we can find several advantages of this method compared to the previous one which uses fixed similarity matrix(+1 for match, -1 for mismatch, -2 for gap) and also can find that the adaptive similarity matrix can be used for detecting various plagiarism cases.

• Archives of Plastic Surgery
• /
• v.34 no.1
• /
• pp.70-76
• /
• 2007

Purpose: Full-thickness skin grafts are usually used in facial reconstruction, but on occasion, split-thickness skin graft is also used from the scalp due to the limitation of donor site. However, there were complications, such as alopecia, folliculitis and blood loss. In addition, it can not be used in patients with baldness. Under the circumstances, we used medial arm skin as split-thickness skin graft donor site in lieu of scalp. We investigated the efficacy of the medial arm skin as a donor site of facial skin graft in comparison with scalp. Methods: From 2000 to 2005, the split-thicknesss skin grafts were performed using the medial arm skin in 10 patients and the scalp in 10 patients. We inspected the skin color match, texture match by the visual analogue scale. Scar contracture was estimated by the Visitrak $grade^{(R)}$(Smith & Nephew). The statistical analysis was performed by SPSS 12.0. Results: There was a more satisfaction in color match, texture, and scar contracture in medial arm skin than in scalp. Conclusion: According to these results, medial arm skin may be used efficiently as an alternative donor site of scalp in the facial reconstruction.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.213-221
• /
• 2021

In this paper, we suggest the Deep Neural Network Model System for predicting results of the match of 'League of Legends (LOL).' The model utilized approximately 26,000 matches of the LOL game and Keras of Tensorflow. It performed an accuracy of 93.75% without overfitting disadvantage in predicting the '2020 League of Legends Worlds Championship' utilizing the real data in the middle of the game. It employed functions of Sigmoid, Relu and Logcosh, for better performance. The experiments found that the four variables largely affected the accuracy of predicting the match --- 'Dragon Gap', 'Level Gap', 'Blue Rift Heralds', and 'Tower Kills Gap,' and ordinary users can also use the model to help develop game strategies by focusing on four elements. Furthermore, the model can be applied to predicting the match of E-sports professional leagues around the world and to the useful training indicators for professional teams, contributing to vitalization of E-sports.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.177-179
• /
• 2022

In the Google Deepmind Challenge match, Alphago defeated Korea's Sedol Lee (human) with 4 wins and 1 loss in the Go match. Finally, artificial intelligence is going beyond the use of human intelligence. The Korean government's budget for the Digital New Deal is 9 trillion won in 2022, and an additional 301 types of data construction projects for artificial intelligence learning will be secured. From 2023, the industrial paradigm will change with the use and application of learning of artificial intelligence in all fields of industry. This paper conducts research to utilize artificial intelligence algorithms. Focusing on the analysis and judgment of data in artificial intelligence learning, research on the appropriate target and scope of application of algorithms in artificial intelligence machine learning and deep learning learning is conducted. This study will provide basic data for artificial intelligence in the 4th industrial revolution technology and artificial intelligence robot use in the 5th industrial revolution technology.

• The Journal of the Korea Contents Association
• /
• v.14 no.5
• /
• pp.390-402
• /
• 2014

This study, based upon the Sturges' crisis stages, examines the crisis response communication strategies of Korea Professional Football League(KPFL) for the K-league match-fixing case as well as the frequency of related news articles and the source of information. To explore the crisis response communication strategies the Korea Professional Football League used, a total of 118 news articles were analyzed using the content analysis and frequency analysis. The unit of analysis for crisis response communication strategies is sentence. The frequency of news articles based upon the crisis stages shows highest rate at the acute crisis stage. The source of information for news reports shows that KPFL was one of the major sources of the news reports. KPFL's crisis response communication strategy throughout all stages of the crisis stage shows that corrective action strategy was used highest ratio. In particular, the crisis response communication strategy between team, player and KPFL was shown lack of consistency throughout all crisis stages. Implication and future research direction for the results are discussed.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.11
• /
• pp.109-115
• /
• 1994

In this paper, we developed an analyzing method of the motional characteristics of sperm, using image processing technology. Without the aid of a dedicated image-processor, this processing of a personal computer(PC) and a simple image processing board. The image processing board is used for acquiring images from a microscopic imaging source. The PC processes the images from the board and computes the parameters of motional characteristics of sperms. The algorithm of the site detection of sperms and the 'Match Matrix Method' is noteworthy. After comparing the results of our method with those of the manual method, and with those of the method using a dedicated image-processor, we concluded that our method is useful and reliable.

• Phonetics and Speech Sciences
• /
• v.8 no.4
• /
• pp.49-61
• /
• 2016

The present study investigates the perception of Korean phonation types by native speakers of Aymara. Perception tests were conducted on two sets of Korean speech materials to determine correspondence between Korean and Aymara 3-way contrasts and to find out which of the consonantal and vocalic part of the syllable is more influential in the perception of Korean phonation types. A set of manipulated stimuli, as well as a set of 12 spontaneous words, were prepared for the tests. The first syllable of the 12 Korean bisyllabic words of 3 series of phonation types(Lenis, Aspirated, and Fortis) in 4 places of articulation were split into consonantal and vocalic parts. And then the two parts were combined to form 9 tokens of CV sequences respectively for each place of articulation. Native speakers of Aymara were forced to match Korean stimuli with one of the 15 Aymara words which represent 3 series of consonant types(plain, aspirated, and ejective) in 5 places of articulation(bilabial, alveolar, palatal, velar, and uvular). Results showed that the consonantal part is more influential than the vocalic part to the Aymara subjects' perception of Korean phonation types when the consonantal part is Aspirated in its phonation type, but the vocalic part is more influential than the consonantal part when the consonantal part is Lenis or Fortis in its phonation type. Response analysis showed that Aymara subjects tend to match Korean stops to Aymara ones in such a way that Lenis corresponds to aspirated, Aspirated to aspirated, and Fortis to plain.

• Journal of the Korean earth science society
• /
• v.26 no.2
• /
• pp.129-136
• /
• 2005

Continued observation of ARGO floats for years(about 4 years) makes the conductivity sensor more vulnerable to fouling by marine life and associated drift in salinity measurements. In this paper, we address this issue by making use of floats deployed in different years. Floats deployed in the East Sea and the Indian Ocean are examined to find out float-to-float match-ups in such a way that an older float pops up simultaneously with a newer deployment (with tolerable space-time difference). A time difference of less than five days and space difference of less than 100km are considered for the match-up data sets. For analysis of the salinity drift under the stable water mass, observations of the floats from deepest water masses have been used. From the cross-check of ARGO floats in the East Sea and the Indian Ocean, it is found that there is a systematic drift in the older float compared to later deployments. All drift results, consistently show negative bias indicating the typical nature of drift from fouled sensors. However, the drift is much less than 0.01, the specified accuracy of ARGO program.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.11b
• /
• pp.1044-1048
• /
• 2002

The experimental modal testing has been carried out for the stator of a generator to confirm the vibrational mode shapes and the corresponding natural frequencies. The model of the stator for the vibration analysis was developed and a series of vibration analyses was carried out. And the properties of the solid element were updated to reduce the differences of the natural frequencies between the measured and the analysed. In the vibration anlyses, the axial, radial and circumferential properties of the solid element were separately varied to take into account the orthotropic effect of the laminated structure and to match the primary modes of the stator core which were extracted from the modal testing. After several attempts to match the measured natural frequencies and model shapes, the properties of the stator model were determined. Comparison of the vibration analyses results based on the determined properties showed fairly good coincidence with the measured data.