• Journal of Intelligence and Information Systems
• /
• v.27 no.2
• /
• pp.1-15
• /
• 2021

If it is difficult for us to make decisions, we ask for advice from friends or people around us. When we decide to buy products online, we read anonymous reviews and buy them. With the advent of the Data-driven era, IT technology's development is spilling out many data from individuals to objects. Companies or individuals have accumulated, processed, and analyzed such a large amount of data that they can now make decisions or execute directly using data that used to depend on experts. Nowadays, the recommender system plays a vital role in determining the user's preferences to purchase goods and uses a recommender system to induce clicks on web services (Facebook, Amazon, Netflix, Youtube). For example, Youtube's recommender system, which is used by 1 billion people worldwide every month, includes videos that users like, "like" and videos they watched. Recommended system research is deeply linked to practical business. Therefore, many researchers are interested in building better solutions. Recommender systems use the information obtained from their users to generate recommendations because the development of the provided recommender systems requires information on items that are likely to be preferred by the user. We began to trust patterns and rules derived from data rather than empirical intuition through the recommender systems. The capacity and development of data have led machine learning to develop deep learning. However, such recommender systems are not all solutions. Proceeding with the recommender systems, there should be no scarcity in all data and a sufficient amount. Also, it requires detailed information about the individual. The recommender systems work correctly when these conditions operate. The recommender systems become a complex problem for both consumers and sellers when the interaction log is insufficient. Because the seller's perspective needs to make recommendations at a personal level to the consumer and receive appropriate recommendations with reliable data from the consumer's perspective. In this paper, to improve the accuracy problem for "appropriate recommendation" to consumers, the recommender systems are proposed in combination with context-based deep learning. This research is to combine user-based data to create hybrid Recommender Systems. The hybrid approach developed is not a collaborative type of Recommender Systems, but a collaborative extension that integrates user data with deep learning. Customer review data were used for the data set. Consumers buy products in online shopping malls and then evaluate product reviews. Rating reviews are based on reviews from buyers who have already purchased, giving users confidence before purchasing the product. However, the recommendation system mainly uses scores or ratings rather than reviews to suggest items purchased by many users. In fact, consumer reviews include product opinions and user sentiment that will be spent on evaluation. By incorporating these parts into the study, this paper aims to improve the recommendation system. This study is an algorithm used when individuals have difficulty in selecting an item. Consumer reviews and record patterns made it possible to rely on recommendations appropriately. The algorithm implements a recommendation system through collaborative filtering. This study's predictive accuracy is measured by Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). Netflix is strategically using the referral system in its programs through competitions that reduce RMSE every year, making fair use of predictive accuracy. Research on hybrid recommender systems combining the NLP approach for personalization recommender systems, deep learning base, etc. has been increasing. Among NLP studies, sentiment analysis began to take shape in the mid-2000s as user review data increased. Sentiment analysis is a text classification task based on machine learning. The machine learning-based sentiment analysis has a disadvantage in that it is difficult to identify the review's information expression because it is challenging to consider the text's characteristics. In this study, we propose a deep learning recommender system that utilizes BERT's sentiment analysis by minimizing the disadvantages of machine learning. This study offers a deep learning recommender system that uses BERT's sentiment analysis by reducing the disadvantages of machine learning. The comparison model was performed through a recommender system based on Naive-CF(collaborative filtering), SVD(singular value decomposition)-CF, MF(matrix factorization)-CF, BPR-MF(Bayesian personalized ranking matrix factorization)-CF, LSTM, CNN-LSTM, GRU(Gated Recurrent Units). As a result of the experiment, the recommender system based on BERT was the best.

• Restorative Dentistry and Endodontics
• /
• v.34 no.4
• /
• pp.333-339
• /
• 2009

This study was done to determine if there is any difference in microleakage between experimental composite resins, in which various proportions of three component photoinitiators (Camphoroquinone, OPPI, Amine) were included. Four kinds of experimental composite resin were made by mixing 3.2% silanated barium glass (78 wt.%, average size; 1 ${\mu}m$) with each monomer system including variously proportioned photoinitiator systems used for photoinitiating BisGMA/BisEMA/TEGDMA monomer blend (37.5:37.5:25 wt.%). The weight percentage of each component were as follows (in sequence Camphoroquinone, OPPI, Amine): Group A - 0.5%, 0%, 1% / Group B - 2%, 0.2%, 2% / Group C - 0.2%, 1%, 0.2% / Group D - 1%, 1%, 2%. Each composite resin was used as a filling material for round class V cavities (diameter: 2/3 of mesiodistal width; depth: 1.5 mm) made on extracted human premolars and they were polymerized using curing light unit (XL 2500, 3M ESPE) for 40 s with an intensity of 600 mW/$cm^2$. Teeth were thermocycled fivehundred times between $50^{\circ}C$and $550^{\circ}C$for 30s at each temperature. Electrical conductivity (${\mu}A$) was recorded two times (just after thermocycling and after three-month storage in saline solution) by electrochemical method. Microleakage scores of each group according to evaluation time were as follows [Group: at first record / at second record; unit (${\mu}A$)]: A: 3.80 (0.69) / 13.22 (4.48), B: 3.42 (1.33) / 18.84 (5.53), C: 4.18 (2.55) / 28.08 (7.75), D: 4.12 (1.86) / 7.41 (3.41). Just after thermocycling, there was no difference in microleakage between groups, however, group C showed the largest score after three-month storage. Although there seems to be no difference in microleakage between groups just after thermocycling, composite resin with highly concentrated initiation system or classical design (Camphoroquinone and Amine system) would be more desirable for minimizing microleakage after three-month storage.

• 한국균학회소식:학술대회논문집
• /
• 2014.05a
• /
• pp.13-13
• /
• 2014

Two kinds of mushrooms, Gumji (金芝; Ganoderma) and Soji, were described in old book of Samguksagi (History of the three kingdoms; B.C 57~A.D 668; written by Bu Sik Kim in 1145) in Korea-dynasty. Many kinds of mushrooms were also described in more than 17 kinds of old books during Chosun-dynasty (1392~1910) in Korea. Nowadays, mushroom cultivation has been increased through out the world last decade years. Production of mushrooms has also been increased 10-20% and many varieties have been cultivated. Similar trends were also observed in Korea. Approximately two hundred commercial strains of 37 species in mushrooms were developed and distributed to cultivators. Somatic hybrid variety of oyster mushroom 'Wonhyeong-neutari' were developed by protoplast fusion, and distributed to grower in 1989. The fruiting body yield index of somatic hybrids of Pleurotus ranged between 27 and 155 compared to parental values of 100 and 138. In addition, more diverse mushroom varieties such as Phellinus baumi, Auricularia spp., Pleurotus ferulae, Hericium erinaceus, Hypsizigus marmoreus, Grifola frondosa, Agrocybe aegerita and Pleurotus cornucopiae have been attempted to cultivate in small scale cultivation. Production of mushrooms as food was 190,111 metric tons valued at 800 billion Korean Won (one trillion won if include mushroom factory products; 1dollar = 1,040 Won) in 2011. Major cultivated species are Pleurotus ostreatus, Pleurotus eryngii, Flammulina velutipes, Lentinula edodes, Agaricus bisporus, and Ganoderma lucidum, which cover 90% of total production. Since mushroom export was initiated from 1960 to 1980, the export and import of mushrooms have been increased in Korea. Technology developed for liquid spawn production and automatic cultivation systems lead to the reduction of the production cost resulting in the increasement of mushroom export. However some species were imported because of high production cost for these mushrooms requiring the effective cultivation methods. Developing of effective post-harvest system will be also directly related to mushroom export. In academic area, RDA scientists have been conducting mushroom genome projects. One of the main results is the whole genome sequencing of Flammulina velutipes for molecular breeding. An electrophoretic karyotype of of F. velutipes was obtained using CHEF with 7 chromosomes, with a total genome size of approximately 26.7 Mb. The mususcript of the genome of F. velutipes was published in PLOS ONE this year. For medicinal mushrooms, we have been conducting the genome research on Cordyceps and its related species for developing functional foods using this mushroom. In 2013, Korea Food and Drug Administraion (KFDA) approved Cordyceps mushroom for its value as an immune booster.

• Restorative Dentistry and Endodontics
• /
• v.25 no.1
• /
• pp.123-132
• /
• 2000

To overcome problems of conventional glass ionomers, resin components have been added to glass ionomers. On a continuum between glass ionomers and composites are a variety of blends, employing different proportions of acid-base and free radical reactions to bring about cure. Popular groups defined between the ends are resin-modified glass-ionomers(RMGIs), polyacid-modified composite resins(Compomers) and ionomer modified resins. These groups show different clinical properties, and in selecting these materials for a restoration, one should sufficiently understand these different setting properties. In this study, some difference in the setting characteristics of different groups of hybrid ionomers were examined. Two RMGIs (Fuji2 LC,GC / Vitremer, 3M), three Compomers (Dyract AP, Dentsply / F2000, 3M / Elan, Kerr) were involved in this study. The identification of the setting characteristics of different groups was achieved by a two-stage study. First, thermal analysis was performed by a differential scanning calorimeter, and then the hardness of each group at different depth and time were measured by a micro-hardness tester. Thermal analysis was performed to identify the inorganic filler content and to record the heat change during setting process. The setting process was progressed for each material by chemical set mode and light-cured mode. In the hardness test, samples of materials were prepared with a 6mm-diameter metal ring, and the hardness was measured at the top, and 1mm, 2.5mm, 4mm below at just after a 40 second-cure, and after 10 minutes, 24 hours, and 7 days. Statistical analysis was performed by Mann-Whitney rank sum test to assess significant differences between set modes and types of materials, and by ANOVA and T-test to evaluate the statistical meanings of data at different times and depths of each materials. Followings are findings and conclusions derived from this study. Thermal analysis; 1. Compomers show no evidence of chemical setting while RMGIs exhibit heat output during the process of chemical setting. 2. Heat of cure of RMGIs exceed Compomers. 3. The net heat output of RMGIs through light-cured mode is higher than through chemically set mode. Hardness test; 1. Initial hardness of RMGIs immediately after light cure is relatively low, but the hardness increases as time goes by. On the contrary, Comomers do not show evident increase of the hardness following time. 2. Compomers show a marked decrease of setting degree as the depth of the material increases. In RMGIs, the setting degree at different depths does not significantly differ.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.5B
• /
• pp.397-408
• /
• 2009

In this study, a stochastic precipitation generation framework for simultaneous simulation of daily precipitation at multiple sites is presented. The precipitation occurrence at individual sites is generated using hybrid-order Markov chain model which allows higher-order dependence for dry sequences. The precipitation amounts are reproduced using Anscombe residuals and gamma distributions. Multisite spatial correlations in the precipitation occurrence and amount series are represented with spatially correlated random numbers. The proposed model is applied for a network of 17 locations in the middle of Korean peninsular. Evaluation statistics are reported by generating 50 realizations of the precipitation of length equal to the observed record. The analysis of results show that the model reproduces wet day number, wet and dry day spell, and mean and standard deviation of wet day amount fairly well. However, mean values of 50 realizations of generated precipitation series yield around 23% Root Mean Square Errors (RMSE) of the average value of observed maximum numbers of consecutive wet and dry days and 17% RMSE of the average value of observed annual maximum precipitations for return periods of 100 and 200 years. The provided model also reproduces spatial correlations in observed precipitation occurrence and amount series accurately.

• Korean Journal of Plant Taxonomy
• /
• v.39 no.1
• /
• pp.12-23
• /
• 2009

To verify hybridity of Asplenium castaneo-viride, external morphology, spore morphology, anatomy and chromosomes of the species and of the two presumed parental species, A. incisum and A. ruprechtii, were examined. A. castaneo-viride usually had 1-pinnately divided frond. However, some individuals had almost simple fronds with pinnatisect basal parts similar to A. ruprechtii, while others had fronds similar to A. incisum in having oblanceolate blades and basal pinnae with triangular, 2-3 lobed apices. On the surface of the spores, sculpturing consisted of folds that were usually prominent; forming long wings, and irregular or incomplete reticulation. However, reticulation patterns varied among species. A. castaneo-viride showed a wide range of variation from sparse to dense patterns, whereas A. incisum showed only from sparse to intermediate patterns. A. ruprechtii showed from intermediate to dense patterns. The spore size of A. castaneo-viride was $54.63{\mu}m$, larger than other two species ($47.81{\mu}m$ in A. incisum and $44.22{\mu}m$ in A. ruprechtii). The level of undulation of epidermal cell wall was also different. A. incisum had the most shallowly undulated wall, and A. castaneo-viride had a pattern intermediate between the two presumed parental species. This same patterns was recognized in the density of stomata. The density of $45.91/mm^2$ in A. castaneo-viride was intermediate between the two presumed parental species ($67.00/mm^2$ in A. incisum, and $37.86/mm^2$ in A. ruprechtii). Chromosome number was constant (2x =2n = 72) as in A. incisum and A. ruprechtii. However, A. castaneo-viride showed a different ploidy level. The populations of Mt. Mai (Jeonbuk province) and Mt. Duryun (Jeonnam province) were diploid (2n = 72) which is a new record for this taxon, whereas the population of Mt. Buram (Seoul) was tetraploid (2n = 144). Conclusively, A. castaneo-viride was revealed to be a hybrid of A. ruprechtii and A. incisum based on evidence involving leaves, spores, epidermal cells, stomata and chromosome number.

• Journal of Sericultural and Entomological Science
• /
• v.49 no.1
• /
• pp.14-17
• /
• 2007

A new silkworm variety Golden silk for spring rearing season is $F_1$ hybrid between Jam311, a Japanese race bred from introduction breeding and Jam312, a Chinese race from introduction breeding. In the local adaptability test performed at 8 local areas in spring of 2006, "Golden silk" was showed 7 percent higher in hatchability than authorized silk-worm races laval. "Golden silk" was receded higher 1.7 percent, 2.2 percent in pupation rate, best cocoon rate respectively than Chunsujam. The larval period shortened for 1.12 day even though low record to the several items including productivity than Chunsujam. "Golden silk" was showed 1,031 m, 2.97 denier in length of cocoon filament, weight of cocoon filament respectively. The concentration of Deoxynojirimycin(DNJ) was measured 4.34 mg in "Golden silk", high then to "Chunsujam"(3.81 mg). In the Paecilomyces tenuipes production ability test, the pupal weight of "Golden silk" was 1.20g.