• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.

• Journal of the Daesoon Academy of Sciences
• /
• v.17
• /
• pp.197-225
• /
• 2004

The respective history of the Chinese Ancient Dynasties from the era of the ancient Emperors to the revolutionary era of Yin-Zhou殷周 was described in the Rongchengshi容成氏, one of the texts in Shanghai Bowuguan zang Zhanguo Chuzhushu(II)上海博物館藏戰國楚竹書(二) discovered in 1994 at an antique market in Hong Kong. Drawn from the historical explanations expressed in it, the anonymous author's own views on history could be observed as largely being composed of 'resignation'禪讓, 'usurpation'簒奪, and 'banishment/smite'放伐. Following the advent of the recently excavated bamboo slips of Rongchengshi, a careful reconsideration is urgently needed to the established interpretation on the origin of the ritual of resignation in the relevant academic circles. Because it shows us that the ritual of resignation as a way of the succession, judging from my analysis, was already realized by Nine Emperors('Rongchengshi', Zunlushi尊盧氏, Hexushi赫胥氏, Gaoxinshi高辛氏, Cangjieshi倉頡氏, Xuanyuanshi軒轅氏, Shennongshi神農氏, 渾沌氏, and Baoxishi包羲氏 and maybe the more) before the era of 'Yao-Shun'堯舜. Accordingly, the aforementioned fact, which has never been elaborated in the previous texts including the first Chinese historiography Shiji史記, is the only peculiar feature to the Rongchengshi itself. Thus, a simple but empirically important question could be raised here: Was this way of description an exceptional case, even as unaccepted in Warring States Period at that time? If then, the Rongchengshi could not but help being evaluated merely as a buried historical texts, without any influence on the ancient Chinese, along with its author. The Chu bamboo slips Tangyu zhi dao唐虞之道 from Guodian Chujian郭店 excavated in 1993, however, has a very similar content to the Rongchengshi in relevance to the historical existence of the ritual of resignation. From the sentence, expressed in Tangyu zhi dao, that "the sudden rise of 'Six Emperors'六帝 was due to the practice of resignation like the period of Yao-Shun", it could be easily presumed that the 'Six Emperors' was closely connected to the 'Nine Emperors' and maybe the more at that time. Comparing with the related extant literary texts and the excavated materials in a vigorous way, in this paper, I explore four significant questions from a more critical stand to the conventional studies. First of all, I explicate the distinctiveness of the Rongchengshi as an academically very precious materials. Secondly, and closely related to the above, I evaluate its status or significance in the history of the Chinese ancient thoughts. And thirdly, I make an endeavor to trace back to the date of its transcription. Finally, and the most crucially, I attempt to show what Schools it was originated from and what connections it had with the Schools in the acient China. In sum, some concluding remarks, having somewhat insightful and significant implications for the further studies in these issues, could be drawn from my exploration. Viewing from the historical perspective of thoughts, at first, the legend of Ancient Dynasty's Succession before Yao-Shun Era in the Rongchengshi had some interactions directly and closely with Zhangzi莊子, Muzi墨子, Guanzi管子, Xunzi荀子 and Tangyu zhi dao. Also in doing search for the unification in a transitional epoch from the late to the end of the Warring States period, the political stand of Shi士 and Ke客 was reflected in it as well as in the Tangyu zhi dao because they actively wanted to suggest the most appropriate model of the Emperor or the idealistic succession process and political realms.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.25-38
• /
• 2019

Selecting high-quality information that meets the interests and needs of users among the overflowing contents is becoming more important as the generation continues. In the flood of information, efforts to reflect the intention of the user in the search result better are being tried, rather than recognizing the information request as a simple string. Also, large IT companies such as Google and Microsoft focus on developing knowledge-based technologies including search engines which provide users with satisfaction and convenience. Especially, the finance is one of the fields expected to have the usefulness and potential of text data analysis because it's constantly generating new information, and the earlier the information is, the more valuable it is. Automatic knowledge extraction can be effective in areas where information flow is vast, such as financial sector, and new information continues to emerge. However, there are several practical difficulties faced by automatic knowledge extraction. First, there are difficulties in making corpus from different fields with same algorithm, and it is difficult to extract good quality triple. Second, it becomes more difficult to produce labeled text data by people if the extent and scope of knowledge increases and patterns are constantly updated. Third, performance evaluation is difficult due to the characteristics of unsupervised learning. Finally, problem definition for automatic knowledge extraction is not easy because of ambiguous conceptual characteristics of knowledge. So, in order to overcome limits described above and improve the semantic performance of stock-related information searching, this study attempts to extract the knowledge entity by using neural tensor network and evaluate the performance of them. Different from other references, the purpose of this study is to extract knowledge entity which is related to individual stock items. Various but relatively simple data processing methods are applied in the presented model to solve the problems of previous researches and to enhance the effectiveness of the model. From these processes, this study has the following three significances. First, A practical and simple automatic knowledge extraction method that can be applied. Second, the possibility of performance evaluation is presented through simple problem definition. Finally, the expressiveness of the knowledge increased by generating input data on a sentence basis without complex morphological analysis. The results of the empirical analysis and objective performance evaluation method are also presented. The empirical study to confirm the usefulness of the presented model, experts' reports about individual 30 stocks which are top 30 items based on frequency of publication from May 30, 2017 to May 21, 2018 are used. the total number of reports are 5,600, and 3,074 reports, which accounts about 55% of the total, is designated as a training set, and other 45% of reports are designated as a testing set. Before constructing the model, all reports of a training set are classified by stocks, and their entities are extracted using named entity recognition tool which is the KKMA. for each stocks, top 100 entities based on appearance frequency are selected, and become vectorized using one-hot encoding. After that, by using neural tensor network, the same number of score functions as stocks are trained. Thus, if a new entity from a testing set appears, we can try to calculate the score by putting it into every single score function, and the stock of the function with the highest score is predicted as the related item with the entity. To evaluate presented models, we confirm prediction power and determining whether the score functions are well constructed by calculating hit ratio for all reports of testing set. As a result of the empirical study, the presented model shows 69.3% hit accuracy for testing set which consists of 2,526 reports. this hit ratio is meaningfully high despite of some constraints for conducting research. Looking at the prediction performance of the model for each stocks, only 3 stocks, which are LG ELECTRONICS, KiaMtr, and Mando, show extremely low performance than average. this result maybe due to the interference effect with other similar items and generation of new knowledge. In this paper, we propose a methodology to find out key entities or their combinations which are necessary to search related information in accordance with the user's investment intention. Graph data is generated by using only the named entity recognition tool and applied to the neural tensor network without learning corpus or word vectors for the field. From the empirical test, we confirm the effectiveness of the presented model as described above. However, there also exist some limits and things to complement. Representatively, the phenomenon that the model performance is especially bad for only some stocks shows the need for further researches. Finally, through the empirical study, we confirmed that the learning method presented in this study can be used for the purpose of matching the new text information semantically with the related stocks.

• MISULJARYO - National Museum of Korea Art Journal
• /
• v.98
• /
• pp.22-53
• /
• 2020

This paper analyzes the contents, characteristics, and historical significance of the dedicatory inscriptions (josanggi) on the Amitabha Buddha and the Maitreya Bodhisattva statues of Gamsansa Temple, two masterpieces of Buddhist sculpture from the Unified Silla period. In the first section, I summarize research results from the past century (divided into four periods), before presenting a new perspective and methodology that questions the pre-existing notion that the Maitreya Bodhisattva has a higher rank than the Amitabha Buddha. In the second section, through my own analysis of the dedicatory inscriptions, arrangement, and overall appearance of the two images, I assert that the Amitabha Buddha sculpture actually held a higher rank and greater significance than the Maitreya Bodhisattva sculpture. In the third section, for the first time, I provide a new interpretation of two previously undeciphered characters from the inscriptions. In addition, by comparing the sentence structures from the respective inscriptions and revising the current understanding of the author (chanja) and calligrapher (seoja), I elucidate the possible meaning of some ambiguous phrases. Finally, in the fourth section, I reexamine the content of both inscriptions, differentiating between the parts relating to the patron (josangju), the dedication (josang), and the prayers of the patrons or donors (balwon). In particular, I argue that the phrase "for my deceased parents" is not merely a general axiom, but a specific reference. To summarize, the dedicatory inscriptions can be interpreted as follows: when Kim Jiseong's parents died, they were cremated and he scattered most of their remains by the East Sea. But years later, he regretted having no physical memorial of them to which to pay his respects. Thus, in his later years, he donated his estate on Gamsan as alms and led the construction of Gamsansa Temple. He then commissioned the production of the two stone sculptures of Amitabha Buddha and Maitreya Bodhisattva for the temple, asking that they be sculpted realistically to reflect the actual appearance of his parents. Finally, he enshrined the remains of his parents in the sculptures through the hole in the back of the head (jeonghyeol). The Maitreya Bodhisattva is a standing image with a nirmanakaya, or "transformation Buddha," on the crown. As various art historians have pointed out, this iconography is virtually unprecedented among Maitreya images in East Asian Buddhist sculpture, leading some to speculate that the standing image is actually the Avalokitesvara. However, anyone who reads the dedicatory inscription can have no doubt that this image is in fact the Maitreya. To ensure that the sculpture properly embodied his mother (who wished to be reborn in Tushita Heaven with Maitreya Bodhisattva), Kim Jiseong combined the iconography of the Maitreya and Avalokitesvara (the reincarnation of compassion). Hence, Kim Jiseong's deep love for his mother motivated him to modify the conventional iconography of the Maitreya and Avalokitesvara. A similar sentiment can be found in the sculpture of Amitabha Buddha. To this day, any visitor to the temple who first looks at the sculptures from the front before reading the text on the back will be deeply touched by the filial love of Kim Jiseong, who truly cherished the memory of his parents.