• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.47-56
• /
• 2024

This paper proposes a method for fine-tuning BERT-based natural language processing models to perform sentiment analysis on Korean review data. By varying the input sequence length during this process and comparing the performance, we aim to explore the optimal performance according to the input sequence length. For this purpose, text review data collected from the clothing shopping platform M was utilized. Through web scraping, review data was collected. During the data preprocessing stage, positive and negative satisfaction scores were recalibrated to improve the accuracy of the analysis. Specifically, the GPT-4 API was used to reset the labels to reflect the actual sentiment of the review texts, and data imbalance issues were addressed by adjusting the data to 6:4 ratio. The reviews on the clothing shopping platform averaged about 12 tokens in length, and to provide the optimal model suitable for this, five BERT-based pre-trained models were used in the modeling stage, focusing on input sequence length and memory usage for performance comparison. The experimental results indicated that an input sequence length of 64 generally exhibited the most appropriate performance and memory usage. In particular, the KcELECTRA model showed optimal performance and memory usage at an input sequence length of 64, achieving higher than 92% accuracy and reliability in sentiment analysis of Korean review data. Furthermore, by utilizing BERTopic, we provide a Korean review sentiment analysis process that classifies new incoming review data by category and extracts sentiment scores for each category using the final constructed model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.6
• /
• pp.1183-1190
• /
• 2017

With the development of network and IT technology, people are searching and purchasing items they want, not bounded by places. Therefore, there are various studies on how to solve the scalability problem due to the rapidly increasing data in the recommendation system. In this paper, we propose an item-based collaborative filtering method using Tag weight and a recommendation technique using MapReduce method, which is a distributed parallel processing method. In order to improve speed and efficiency, the proposed method classifies items into categories in the preprocessing and groups according to the number of nodes. In each distributed node, data is processed by going through Map-Reduce step 4 times. In order to recommend better items to users, item tag weight is used in the similarity calculation. The experiment result indicated that the proposed method has been more enhanced the appropriacy compared to item-based method, and run efficiently on the large amounts of data.

• Journal of Korean Society of Transportation
• /
• v.24 no.3 s.89
• /
• pp.7-15
• /
• 2006

A large sample can Provide more information about the Population. As the sample size Increases, analysts will be more confident about the survey results. On the other hand, the costs for survey will increase in time and manpower. Therefore, determination of the sample size is a trade-off between the required accuracy and the cost. In addition, permitted error and significance level should be considered. Sample size determination in surveys for O/D estimation is also connected with confidence of survey result. However, the past methods were usually too simple to consider confidence. Therefore, a new method for O/D surveys was Proposed and it was accurate enough, but it has too large sample size when we have current budget constraint. In this research, several minimum sample size determination methods for origin-destination survey under budget constraint were proposed. Each method decreased sample size, but has its own advantages. Selection of the sample size will depend on the study Purpose and budget constraint.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.2
• /
• pp.453-460
• /
• 2013

This study proposes a framework that enables efficient integration and usage of enterprise data using semantic based concept net. Integration of enterprise information that has been increasing geometrically in cloud environment. The concept net is very similar in approaching way to existing ontology. However, it builds correlation between object and concept to help user's information integration retrieval more efficiently. In this study, concept nets are divided into 3 kinds and are applied to the proposed framework independently. The concept net in this study is built in ontology format based on master information concept net, keyword concept net and business process concept net. This concept net enables retrieval and usage of data based on correlation among data according to user's request. Then, through combination of master information concept and keyword concept, it provides frequency trace of keyword and category thus improving convenience and speed of retrieval.

• Journal of Internet Computing and Services
• /
• v.6 no.5
• /
• pp.143-154
• /
• 2005

This paper presents a reliable multicast transmission for the advertising-supported Access Point in which a user can use a wireless network access service through receiving the advertisement. In this paper we propose a application-layer multicast protocol that controls a transmission rate of the mobile device for the reliable multicast in wireless LAN environment. Internet advertising includes all means and medias for advertising on the Internet in order to raise sales or popularity of the products or services. Since the current Internet advertising systems are passive, the target systems are exposed to unspecified persons and its exposure rates of the advertisement are changeable and unpredictable. In this paper, we propose an Internet advertising system, with which users can access the wireless Internet without charge, advertisers can provide customized advertisement according to location, time, and categories of users, and owners of network infrastructure can manage the system with a low cost.

• Archives of design research
• /
• v.15 no.3
• /
• pp.167-180
• /
• 2002

This study investigates effective ways to help users form a correct mental model of the hierarchical information space (HIS) in small display. The focus is the effect of color cue on understanding the structure and navigating the information space. The concept of color space (CS) corresponds well to the HIS - one color has a unique position in the CS as a piece of information does in HIS. In this study, we empirically examined two types of color cue, namely, categorical and depth cue. Hue was used as a categorical cue and tone was used as a depth cue. In our experiment, we evaluate the effectiveness of the color cues in the mobile internet system. Subjects were asked to perform four searching tasks and four comparison tasks. The results of experiment reveal that the categorical cues significantly improve the user's mental model whereas decrease navigation performances. The depth cues cannot aid in understanding the HIS as well as improve navigation performances. This study concludes with limitations of the study and descriptions of future studies.

• The Journal of Society for e-Business Studies
• /
• v.26 no.3
• /
• pp.67-80
• /
• 2021

The node embedding technique for learning graph representation plays an important role in obtaining good quality results in graph mining. Until now, representative node embedding techniques have been studied for homogeneous graphs, and thus it is difficult to learn knowledge graphs with unique meanings for each edge. To resolve this problem, the conventional Triple2Vec technique builds an embedding model by learning a triple graph having a node pair and an edge of the knowledge graph as one node. However, the Triple2 Vec embedding model has limitations in improving performance because it calculates the relationship between triple nodes as a simple measure. Therefore, this paper proposes a feature extraction technique based on a graph convolutional neural network to improve the Triple2Vec embedding model. The proposed method extracts the neighborliness vector of the triple graph and learns the relationship between neighboring nodes for each node in the triple graph. We proves that the embedding model applying the proposed method is superior to the existing Triple2Vec model through category classification experiments using DBLP, DBpedia, and IMDB datasets.

• Smart Media Journal
• /
• v.10 no.2
• /
• pp.70-75
• /
• 2021

Grouping and classifying similar designs in design increase efficiency in terms of management and provide convenience in terms of use. Using artificial intelligence algorithms, this study attempted to classify textile designs into four categories: dots, flower patterns, stripes, and geometry. In particular, we explored whether it is possible to find and explain the regions of interest underlying classification from the perspective of artificial intelligence. We randomly extracted a total of 4,536 designs at a ratio of 8:2, comprising 3,629 for training and 907 for testing. The models used in the classification were VGG-16 and ResNet-34, both of which showed excellent classification performance with precision on flower pattern designs of 0.79%, 0.89% and recall of 0.95% and 0.38%. Analysis using the Local Interpretable Model-agnostic Explanation (LIME) technique has shown that geometry and flower-patterned designs derived shapes and petals from the region of interest on which classification was based.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.2
• /
• pp.238-243
• /
• 2007

The TAM(Topographic Attentive Mapping) network neural network model is an especially effective one for pattern analysis. It is composed of of Input layer, category layer, and output layer. Fuzzy rule, lot input and output data are acquired from it. The TAM network with three pruning rules for reducing links and nodes at the layer is called fuzzy TAM network. In this paper, we apply fuzzy TAM network to pattern analysis of leadership type for organizational leader and show its usefulness. Here, criteria of input layer and target value of output layer are the value and leadership related personality type variables of the Egogram and Enneagram, respectively.

• The Research Journal of the Costume Culture
• /
• v.21 no.3
• /
• pp.338-347
• /
• 2013

In this study, it was explored a research model consisting of slogan attitude, brand familiarity, brand attitude, brand recall, and product category recall of retailers. Experimental research was conducted with 3,028 males and females in their 20's to 40's using stimuli of 10 slogan-brand sets from various types of retailers. In results, the research model developed based on the literature was confirmed and supported by data. In the model test, all hypotheses were supported. The effects of slogan attitude and brand familiarity on brand attitude were confirmed. Also, brand familiarity affected brand recall. Category recall was predicted by brand attitude and brand recall. As consumers have better attitude toward slogans, they tend to have better attitude toward the brand. As consumers are more familiar with the brand, they are likely to better recall brands when they are exposed to the slogan. As consumers have better attitude toward brand and better recall the brand, they tend to better recall the business category when they see the slogan. Study findings may help marketers to develop better strategies for slogan use by considering diverse variables related to consumer responses toward slogan attitudes.