• Journal of The Korean Society of Grassland and Forage Science
• /
• v.36 no.4
• /
• pp.344-349
• /
• 2016

This study was conducted to evaluate the possibility of expanding the usage of whole crop silage from beef cattle and dairy cow to hogs and chickens. For this purpose, a crushing device was developed to crush whole crop silage. The crushed silage was sealed, and analyzed for its feed value. The silage varieties used for the experiment included Saessal barley and Geumgang wheat. Whole crop barley and wheat were crushed in the crushing system as a whole without separating stems, leaves, grains, etc.. When the crushed whole crop silages (CWCS) were analyzed, full grain, grains above 10 mm in size, grains 5~10 mm in size, and grains below 5 mm in size accounted for, 20%, 4%, 27%, and 49 %, respectively. In order to facilitate the fermentation of CWCS, inoculated some fermenter into each CWCS sample (barley or wheat). As control, another set of sample was not inoculated. Crude protein (CP), ether extract (EE), crude fiber (CF), neutral detergent fiber (NDF), acid detergent fiber (ADF), lignin, cellulose content, total digestible nutrient (TDN), and relative feed value (RFV) of fermenter-inoculated Saessal barley were 2.45 %, 1.61%, 8.95%, 16.94%, 9.52%, 1.01%, 8.51%, 81.38%, and 447.5%, respectively. The CP, EE, CF, NDF, ADF, lignin, cellulose content, TDN, and RFV in the other sample of Saessal barley without inoculation of fermenter were 2.57%, 1.62%, 9.61%, 18.25%, 10.13%, 1.10%, 9.04%, 80.90%, and 412.9%, respectively. The CP, EE, CF, NDF, ADF, lignin, cellulose content, TDN, and RFV of fermenter-inoculated Geumgang wheat sample were 2.43%, 1.27%, 10.99%, 19.49%, 11.23%, 1.46%, 9.77%, 80.03%, and 382.6%, respectively. The CP, EE, CF, NDF, ADF, lignin, cellulose content, TDN, RFV of the other set sample of Geumgang wheat sample without the inoculation of fermenter were 2.28%, 1.44%, 10.08%, 18.02%, 10.44%, 1.26%, 9.18%, 80.65%, and 416.9%, respectively. The TDN and RFV content in the fermenter-inoculated Saessal barley were 81.38% and 447.5%, respectively, while the one in the fermenter-inoculated Geumgang wheat were 80.03% and 382.6% respectively. When the feed value of whole crop barley and wheat silage without crushing process was compared to the feed value of whole crop barley and wheat silage made from crushing system, the latter appeared to be higher than the former. This could be due to the process of sealing the crushed silage which might have minimized air content between samples and shortened the golden period of fermentation. In conclusion, these results indicate that a crushing process might be needed to facilitate fermentation and improve the quality of silage when making whole crop silage.

• Journal of Intelligence and Information Systems
• /
• v.21 no.4
• /
• pp.79-92
• /
• 2015

Knowledge map is widely used to represent knowledge in many domains. This paper presents a method of integrating the national R&D data and assists of users to navigate the integrated data via using a knowledge map service. The knowledge map service is built by using a lightweight ontology and a topic modeling method. The national R&D data is integrated with the research project as its center, i.e., the other R&D data such as research papers, patents, and reports are connected with the research project as its outputs. The lightweight ontology is used to represent the simple relationships between the integrated data such as project-outputs relationships, document-author relationships, and document-topic relationships. Knowledge map enables us to infer further relationships such as co-author and co-topic relationships. To extract the relationships between the integrated data, a Relational Data-to-Triples transformer is implemented. Also, a topic modeling approach is introduced to extract the document-topic relationships. A triple store is used to manage and process the ontology data while preserving the network characteristics of knowledge map service. Knowledge map can be divided into two types: one is a knowledge map used in the area of knowledge management to store, manage and process the organizations' data as knowledge, the other is a knowledge map for analyzing and representing knowledge extracted from the science & technology documents. This research focuses on the latter one. In this research, a knowledge map service is introduced for integrating the national R&D data obtained from National Digital Science Library (NDSL) and National Science & Technology Information Service (NTIS), which are two major repository and service of national R&D data servicing in Korea. A lightweight ontology is used to design and build a knowledge map. Using the lightweight ontology enables us to represent and process knowledge as a simple network and it fits in with the knowledge navigation and visualization characteristics of the knowledge map. The lightweight ontology is used to represent the entities and their relationships in the knowledge maps, and an ontology repository is created to store and process the ontology. In the ontologies, researchers are implicitly connected by the national R&D data as the author relationships and the performer relationships. A knowledge map for displaying researchers' network is created, and the researchers' network is created by the co-authoring relationships of the national R&D documents and the co-participation relationships of the national R&D projects. To sum up, a knowledge map-service system based on topic modeling and ontology is introduced for processing knowledge about the national R&D data such as research projects, papers, patent, project reports, and Global Trends Briefing (GTB) data. The system has goals 1) to integrate the national R&D data obtained from NDSL and NTIS, 2) to provide a semantic & topic based information search on the integrated data, and 3) to provide a knowledge map services based on the semantic analysis and knowledge processing. The S&T information such as research papers, research reports, patents and GTB are daily updated from NDSL, and the R&D projects information including their participants and output information are updated from the NTIS. The S&T information and the national R&D information are obtained and integrated to the integrated database. Knowledge base is constructed by transforming the relational data into triples referencing R&D ontology. In addition, a topic modeling method is employed to extract the relationships between the S&T documents and topic keyword/s representing the documents. The topic modeling approach enables us to extract the relationships and topic keyword/s based on the semantics, not based on the simple keyword/s. Lastly, we show an experiment on the construction of the integrated knowledge base using the lightweight ontology and topic modeling, and the knowledge map services created based on the knowledge base are also introduced.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.26 no.2
• /
• pp.377-398
• /
• 1999

Calcium fluoride, created by topical fluoride application, is the reservoir for fluoride ion regulated by pH in the oral environment. Therefore, the amount and the maintenance of calcium fluoride have an important role in preventing dental caries. The aim of this study is to evaluate the effect of Nd:YAG laser irradiation on the generation of calcium fluoride and the acid resistance of tooth enamel. The bovine anterior permanent teeth were prepared (n=276), and divided into following groups : no treatment (control) fluoride application alone, laser irradiation alone, laser irradiation after fluoride application, and fluoride application after laser irradiation. And each group was subdivided based on the application time of 1.23% acidulated phosphate fluoride (APF) (5 min and 30 min) and the irradiation energy of Nd:YAG laser ($20J/cm^2\;and\;40J/cm^2$). In case of fluoride application, each group was divided according to KOH treatment. Twenty three treatment conditions were made for this experiment and twelve specimens were assigned to each treatment condition. In each treatment condition, ten specimens were used for chemical analysis and two specimens were observed under SEM. In groups without treating KOH, fluoride content and the depth of enamel dissolved were measured using enamel biopsy technique. In groups with treating KOH, the amount of calcium fluoride was measured by the treatment with 1 M KOH for 24 hours and enamel biopsy was performed after KOH treatment. The results were analyzed by the fluoride content and the depth of enamel dissolved by enamel biopsy, amount and thickness of calcium fluoride, and the surface structures of enamel. The results are as follows: 1. In groups without treating KOH, the fluoride content of removed enamel showed a positive relationship with the energy density of laser when the laser irradiated before fluoride application 2. In groups without treating KOH, the depth of enamel dissolved decreased more with the combined laser and fluoride treatment than with laser or fluoride treatment, except for the case of $20J/cm^2$ laser irradiation after 5 minute fluoride application (p<0.05). 3. The amount of calcium fluoride did not increased by laser treatment with no statistical significance(p>0.05). 4. The particle size of calcium fluoride increased in case of fluoride treatment after laser irradiation, compared with fluoride application alone. In case of laser treatment after fluoride application, the particle size of calcium fluoride increased and some of the particles fused as well. 5. There were no significant differences in the fluoride content of dissolved enamel between groups without treating KOH and control group, except for the case of laser irradiation after treatment of APF for 30 minutes (p>0.05). 6. In groups with treating KOH, depth of removed enamel in the groups of combined treatment with laser and fluoride was shallower than that in fluoride application groups (p<0.05). 7. In groups without treating KOH, the relationship between fluoride content and the depth of enamel dissolved showed more negative (Spearman correlation coefficient: -0.6281) than in groups with treating KOH (Spearman correlation coefficient: -0.3792). The greater amount of calcium fluoride could be found in case where there was a significant differences of the depth of enamel dissolved between groups with and without treating KOH. From these results, it can be concluded that laser seems to be a little effects on the amount of calcium fluoride formation, but has some effect on the lowering the solubility of calcium fluoride. As the combined treatment of laser and fluoride application showed more effective acid-resistant property, more extended recall period for fluoride application can be achieved with this combined treatment in the clinic.

• Journal of Korean Society of Forest Science
• /
• v.13 no.1
• /
• pp.67-78
• /
• 1971

The measures of contour-terracing with sod has been executed as a major measures for hillside erosion control works for a long time in Korea. It is, however, recognized that pair terracings make a new slope-face having the more steeper degree of slope between the upper and the lower terraces on hillsides and it also does not contribute for establishing the natural vegetation-cover by penetration of pioneer seeds on the slope faces or cut-faces of hillsides. The study was therefore conducted in connection with the above problems on the cut-face having slope of $40^{\circ}$ and 1.6 meter in slope length with clay soils. Plot allocation for the experiment consists of 3 kinds of 3 replica plots having each $1.6m^2$ of slope area, i. e., the control plot with direct seeding on slopes only ($T_1$), the covering plot with the straw-mats after seeding on slopes ($T_2$) and the seeding plot after covering with the straw-mats. ($T_3$). The main results obtained may be summarized as follows : 1. Effects of the straw-mat mulchings on surface soil loss control:-The total amount of soil losses from each treatments are measured as 4,651 gr from $T_1$, 163 gr. from $T_2$ and 2,891 gr. from $T_3$ treatment respectively. (Refer to table No. 2, 3 and 4). In short, it is recognized that effect of $T_2$ treatment is compared as 28.5 times than that of $T_1$ treatment and 17.7 times than that of $T_3$ treatment respectively. Effect of $T_3$ treatment compared with $T_1$ treatment is also such recognizable as 1.6 times in control of surface soil losses on a slope face. 2. Effect of the straw-mat mulchings on soil moisture content on slopes; -Average per cent of surface soil moisture content by treatments show as 21.60 at the $T_1$, 23.04 at the $T_2$ and 22.21 at the $T_3$ treatment respectively and that of subsurface soil moisture content by treatment show as 23.81 at the $T_1$, 26.16 at the $T_2$ and 24.81 at the $T_3$ treatment respectively. The variance of soil moisture content by treatments was highly significant (Refer table No. 7, 8 and 9). 3. Effect of the straw-mat mulchings on vegetation establishment;-Average numbers of germination by treatments are counted as 237 Nos. at the $T_1$, 246 Nos. at the $T_2$ and 262 Nos. at the $T_3$ treatment plots and the vegetation coverage on ground was almost same as about 90% of covers in all treatments. This effect is more or less lower than that of surface soil erosion control. 4. Regarding the effect on surface soil erosion control, the straw-mat mulchings would be effective as a new measures for control of soil erosion on erosion susceptible lands such slope-faced bare-lands as cut-fill faces, mass-movement faces and bare hillsides.

• Journal of Animal Science and Technology
• /
• v.46 no.3
• /
• pp.373-386
• /
• 2004

This study was conducted to investigate effects of the brown seaweed residues supplementation on in vitro fermentation, and milk yield and milk composition of dairy cows. Therefore, two experiments consisting of an in vitro and an in vivo growth trial were used. In in vitro experiment, brown seaweed residues(BSR) was supplemented in basal diet with 0, 1, 2 and 4% respectively, and incubated for 3, 6, 9, 12, and 24 h. The pH value, ammonia-N and VFA were investigated. The pH value tended to increase with increasing BSR during the incubation. Particularly, pH was significantly higher in BSR treatments compared with control at 9 h(p < 0.05). While, ammonia-N concentration was not significantly different across treatments during the whole incubation. BSR supplementation did not affect total VFA production, but acetate was linearly increased in BSR treatments compared with control at 12 h(p < 0.05), and its concentration was highest(92.70 mM) in 4% BSR among treatments. The concentration of iso-butyrate tended to increase in BSR treatments in comparison to control during the incubation. In addition, the concentration of iso-valerate was higher in BSR treatments compared with control at 12 and 24 h. In growth trial, BSR was added(800 g/d/animaI) to diets of dairy cow. Dry matter intake was not affected by BSR supplementation, but daily milk yield(kg) significantly increased in BSR treatment compared with control(p < 0.05). However, milk composition(%) and milk yield(kg) were not significantly different between treatments. Milk fat(% and kg/d) tended to slightly decrease in BSR treatment compared with control(3.59% and 1.06 kg/d vs. 3.32% and 1.01 kg/d), The contents of C16:0 and C20:4 in milk significantly increased in BSR treatment compared with control reflecting from dietary fatty acid composition. The content of C18:0 in milk which is end product of biohydrogenation of CI8 unsaturated fatty acids in the rumen significantly increased in BSR treatment compared with control(p < 0.05). C18:2 content in milk tended to decrease, but tended to increase trans-II C18:l and CLA contents in milk in BSR treatment compared with control. In conclusion, it could be summarized that BSR may stabilize rumen pH, and it could improve milk yield and CIA content in milk with more than 4% of diet. Therefore, BSR could be beneficially used in dairy diets as a feed additive.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.39 no.3
• /
• pp.197-210
• /
• 2003

The non-float midwater pair trawl was effective in the mouth opening and control of the working depth in midwater and bottom. In contrast, we confirmed that it was difficult to keep the net at surface above 30 m of the depth by means of the full scale experiment in the field and the model test in the circulation water channel. To solve this problem, the kites were attached to the head rope of the non-float midwater pair trawl. In this study, four kinds of the model experiments were carried out with the purpose of applying the kite to the korean midwater pair trawl. The results obtained can be summarized as follows: 1. The working depth of the non-float midwater pair trawl with the kite was shallower than that of the proto type and non-float type. The working depth of the kite type was approximately 20m with 2 kites and about 5m with 4 kites under 4.0 knot. The working depth was almost constant but the depth of the head rope sank approximately 15m and 10m according to the increase in the front weight and the wing-end weight, respectively. The changing aspect of the working depth was constant, but the depth of the head rope sank approximately 22m according to the increase in the lower warp length (dL). 2. The hydrodynamic resistance of the kite type was almost increased in a linear form in accordance with the flow speed increase from 2.0 to 5.0 knot. The increasing grate of the hydrodynamic resistance tended to increase in accordance with the increase in flow speed. The hydrodynamic resistance of the kite type was larger approximately 5～10 ton larger than that of the non-float type and the proto type. The hydrodynamic resistance of the kite type increased approximately 3ton with the changing of the front weight from 1.40 to 3.50 ton and approximately 4 ton with the changing of the wing-end weight from 0 to 1.11 ton and approximately 5.5 ton with the changing lower warp length (dL) from 0 to 40 m, respectively. 3. The net height of the kite type was increased approximately 10 m with the change in the kite area from $2,270mm^2$ to 4,540 $\textrm{mm}^2$. The net height of the kite type was aproximately 50 m and 30 m larger than that of the proto type and the non-float type, respectively. The changed aspect of the net width was approximately 5m with the variation of the flow speed from 2.0 to 5.0 knot. 4. The filtering volume of the kite type was larger than that of the proto type and the non-float type by 28%, 34% at 2.0 knot of the flow speed and 42%, 41% at 3.0 knot, and 62%, 45% at 4.0 knot, and 74%, 54% at 5.0knot, respectively. The optimal towing speed was approximately 3.0 knot for the proto type and was over 4.0 knot for the non-float type, and the optimal towing speed reached 5.0 knot for the kite type. 5. The opening efficiency of the kite type was approximately 50% and 25% larger than that of the proto type and the non-float type, respectively.

• Korean Journal of Poultry Science
• /
• v.35 no.4
• /
• pp.375-380
• /
• 2009

This study was conducted to investigate the effect of feeding induced molting on the visceral organs and blood component profile in laying hens and designed to test 400 flocks of 60 week old Leghorn laying hens for 34 weeks. A total of four molting treatment methods by including the molted with customary molting by fasting method (c), feeding single diet of corn (T1), feeding single diet of wheat bran (T2) and feeding single diet of alfalfa meal (T3) were tested, and each treatment was repeated for 5 times, and 20 laying hens were randomly assigned in an cage for each repeat. As the result of the experiment, ovary was $2.03{\sim}6%$ and oviduct was $2.51{\sim}3.47%$ in visceral organs for body weight at pre-molting term, but there was no significant difference. At post-molting, no significant difference was found, ovary was $0.25{\sim}0.41%$, uterus of control, T1, T2 and T3 was 1.12%, 0.82%, 0.48% and 0.90%, respectively. T2 was significantly lower than control, T3 (p<0.05) at the 50% of egg production. Ovary was $2.20{\sim}2.60%$ and oviduct was $2.98{\sim}3.45%$. In addition, ovary was $2.65{\sim}3.01%$, oviduct was $3.23{\sim}3.64%$ at the peak egg production, but there was no significant difference by non-feeding and feeding molting treatments. In blood component profile, cholesterol was $179.8{\sim}245.7\;mg/dL$ at pre-molting, but there was no significant difference and at post-molting, concentration of cholestrol in control, T1, T2 and T3 was 353.6, 229.1, 261.8 and 300.6 mg/dL, respectively. T1 was significantly lower than control and T3 (p<0.05). In addition, first laying day was $228.1{\sim}271.8\;mg/dL$, 50% of egg production was $236.5{\sim}284.8\;mg/dL$, there was no significant difference. Concentration of cholestrol in control, T1, T2 and T3 was 324.1, 591.6, 363.0 and 315.6 mg/dL, respectively, at the peak egg production period. T1 was significantly higher than other treatment (p<0.05).

• Journal of Chest Surgery
• /
• v.30 no.5
• /
• pp.471-478
• /
• 1997

Introduction: The use of rabbits as a cardiopulmonary bypass(CPB) animal model is extremely dif%cult mainly due to technical problems. On the other hand, deep hypothermic circulatory arrest(CA) is used to facilitate surgical repair in a variety of cardiac diseases. Although steroids are generally known to be effective in the treatment of cerebral edema, the protective effects of steroids on the brain during CA are not conclusively established. Objectives of this study are twofold: the establishment of CPB technique in rabbits and the evaluation of preventive effect of steroid on the development of brain edema during CA. Material '||'&'||' Methods: Fifteen New Zealan white rabbits(average body weight 3.5kg) were divided into three experimental groups; control CA group(n=5), CA with Trendelenberg position group(n=5), and CA with Trendelenberg position + steroid(methylprednisolone 30 mglkg) administration group(n=5). After anesthetic induction and tracheostomy, a median sternotomy was performed. An aortic cannula(3.3mm) and a venous ncannula(14 Fr) were inserted, respectively in the ascending aorta and the right atrium. The CPB circuit consisted of a roller pump and a bubble oxygenator. Priming volume of the circuit was approximately 450m1 with 120" 150ml of blood. CPB was initiated at a flow rate of 80~85ml/kg/min, Ten min after the start of CPB, CA was established with duration of 40min at $20^{\circ}C$ of rectal temperature. After CA, CPB was restarted with 20min period of rewarming. Ten min after weaning, the animal was sacrif;cod. One-to-2g portions of the following tissues were rapidly d:ssected and water contents were examined and compared among gr ups: brain, cervical spinal cord, kidney, duodenum, lung, heart, liver, spleen, pancreas. stomach. Statistical significances were analyzed by Kruskal-Wallis nonparametric test. Results: CPB with CA was successfully performed in all cases. Flow rate of 60-100 mlfkgfmin was able to be maintained throughout CPB. During CPB, no significant metabolic acidosis was detected and aortic pressure ranged between 35-55 mmHg. After weaning from CPB, all hearts resumed normal beating spontaneously. There were no statistically significant differences in the water contents of tissues including brain among the three experimental groups. Conclusion: These results indicate (1) CPB can be reliably administered in rabbits if proper technique is used, (2) the effect of steroid on the protection of brain edema related to Trendelenburg position during CA is not established within the scope of this experiment.

• Journal of Animal Science and Technology
• /
• v.48 no.6
• /
• pp.933-942
• /
• 2006

This experiment was carried out to investigate effects of honeybee venom treatment on the humoral immune response in pigs. Corresponding author : S. K. Cho, Dept. of Animal Sci. Chung-Buk National University, Kaesin-dong, Cheongju, 361-763, Korea. phone : 043-261-2551. E-mail : deercho@chungbuk.ac.kr To investigate effects of natural honeybee venom on the concentration of immunoglobulin G, A, and M, 20 piglets(LY×D) from 3 sows were allocated into two groups bee venom-treated group(10 piglets) and non-treated control(10 piglets). Natural honeybee venom was treated at 0, 3, 6 days after birth and the acupoints were Hai-men(ST-25), Du-kou(CV-8) and Jiao-chao(GV-1) points at 0, 3 days after birth and the regions of castration and tail amputation point at 6 days. Control group was injected 1㎖ of saline to the same site. Concentrations of IgG, A, and M were measured with immunoturbidimetric method at 0, 3, 7, 14, and 21 days after treatment. To investigate the effect of bee venom on the production of antibodies against hog cholera and atrophic rhinitis vaccines that were used as indicator antigens, 40 piglets(LYxD) from 5 sows were grouped as bee venom-treated group (20 piglets) and control group(20 piglets). Natural honeybee venom was treated at 0, 3days(castration, tail amputation) and 21days after birth. The acupoints were Hai-men(ST-25), Du-kou(CV-8) and Jiao-chao (GV-1) points at 0 day, the regions of castration and tail ampution at 3 days and Jiao-chao(GV-1) and Bai-hui(GV-20) points at 21days after birth(weaning). Control group was injected 1ml of saline to the same site. Atrophic rhinitis vaccine was injected twice at 24 and 44 days after birth and hog cholera vaccine was also injected twice at 44 and 64 days after birth. Antibody titers against Bordetella bronchiseptica and hog cholera virus were measured by using tube agglutination and ELISA tests at 24, 34, 44, 54 and 74 days after birth. Concentrations of IgG of treated group were 339.52, 366.48, 296.52, 242.06 and 219.06mg/dl at 0, 3, 7, 14 and 21 days after birth, respectively. In contrast, concentrations of IgG in control group were respectively 347.10, 334.14, 243.28, 205.18 and 191.58mg/dl during same periods with treated group. Concentrations of IgG at 0 day was not significantly different between the treated group and control group but treated group were significantly increased by 10.28% at 3 days after birth (P<0.02), 21.88% at 7 days after birth(P<0.01), 18.0% at 14 days after birth(P<0.07) and 14.3% at 21 days after birth(P<0.01). Concentrations of IgA and Ig M were not significantly different. Antibody titers against hog cholera virus were significantly increased by 57.0% at 24 days after birth(P<0.03), 74.6% at 34 days after birth (P<0.006), 48.6% at 44 days after birth(P<0.017), 45.0% at 54 days after birth(P<0.16) and 44.4% at 74 days after birth (P<0.006) in bee venom treated group in comparison with control group. Antibody titers against the Bordetella bronchiseptica was significantly increased in Beevenom treated group as 9.1% (P<0.32) at 24days, 39.7% (P<0.002) at 34days, 31.9% (P<0.02) at 44days, 33.4% (P<0.01) at 54days and 57.3% (P<0.007) at 74 days after birth when compared with those of control group pigs. Collecting together, the results in this study showed that immune responses were increased by treatment of natural honeybee venom to pigs. These results suggested that the treatment of bee venom could be used effectively for the increase of productivity in livestock industry.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.97-117
• /
• 2020

Product evaluation criteria is an indicator describing attributes or values of products, which enable users or manufacturers measure and understand the products. When companies analyze their products or compare them with competitors, appropriate criteria must be selected for objective evaluation. The criteria should show the features of products that consumers considered when they purchased, used and evaluated the products. However, current evaluation criteria do not reflect different consumers' opinion from product to product. Previous studies tried to used online reviews from e-commerce sites that reflect consumer opinions to extract the features and topics of products and use them as evaluation criteria. However, there is still a limit that they produce irrelevant criteria to products due to extracted or improper words are not refined. To overcome this limitation, this research suggests LDA-k-NN model which extracts possible criteria words from online reviews by using LDA and refines them with k-nearest neighbor. Proposed approach starts with preparation phase, which is constructed with 6 steps. At first, it collects review data from e-commerce websites. Most e-commerce websites classify their selling items by high-level, middle-level, and low-level categories. Review data for preparation phase are gathered from each middle-level category and collapsed later, which is to present single high-level category. Next, nouns, adjectives, adverbs, and verbs are extracted from reviews by getting part of speech information using morpheme analysis module. After preprocessing, words per each topic from review are shown with LDA and only nouns in topic words are chosen as potential words for criteria. Then, words are tagged based on possibility of criteria for each middle-level category. Next, every tagged word is vectorized by pre-trained word embedding model. Finally, k-nearest neighbor case-based approach is used to classify each word with tags. After setting up preparation phase, criteria extraction phase is conducted with low-level categories. This phase starts with crawling reviews in the corresponding low-level category. Same preprocessing as preparation phase is conducted using morpheme analysis module and LDA. Possible criteria words are extracted by getting nouns from the data and vectorized by pre-trained word embedding model. Finally, evaluation criteria are extracted by refining possible criteria words using k-nearest neighbor approach and reference proportion of each word in the words set. To evaluate the performance of the proposed model, an experiment was conducted with review on '11st', one of the biggest e-commerce companies in Korea. Review data were from 'Electronics/Digital' section, one of high-level categories in 11st. For performance evaluation of suggested model, three other models were used for comparing with the suggested model; actual criteria of 11st, a model that extracts nouns by morpheme analysis module and refines them according to word frequency, and a model that extracts nouns from LDA topics and refines them by word frequency. The performance evaluation was set to predict evaluation criteria of 10 low-level categories with the suggested model and 3 models above. Criteria words extracted from each model were combined into a single words set and it was used for survey questionnaires. In the survey, respondents chose every item they consider as appropriate criteria for each category. Each model got its score when chosen words were extracted from that model. The suggested model had higher scores than other models in 8 out of 10 low-level categories. By conducting paired t-tests on scores of each model, we confirmed that the suggested model shows better performance in 26 tests out of 30. In addition, the suggested model was the best model in terms of accuracy. This research proposes evaluation criteria extracting method that combines topic extraction using LDA and refinement with k-nearest neighbor approach. This method overcomes the limits of previous dictionary-based models and frequency-based refinement models. This study can contribute to improve review analysis for deriving business insights in e-commerce market.