• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.31-49
• /
• 2023

With the growth of the food-catering industry, consumer preferences and the number of dine-in restaurants are gradually increasing. Thus, personalized recommendation services are required to select a restaurant suitable for consumer preferences. Previous studies have used questionnaires and star-rating approaches, which do not effectively depict consumer preferences. Online reviews are the most essential sources of information in this regard. However, previous studies have aggregated online reviews into long documents, and traditional machine-learning methods have been applied to these to extract semantic representations; however, such approaches fail to consider the surrounding word or context. Therefore, this study proposes a novel review textual-based restaurant recommendation model (RT-RRM) that uses deep learning to effectively extract consumer preferences from online reviews. The proposed model concatenates consumer-restaurant interactions with the extracted high-level semantic representations and predicts consumer preferences accurately and effectively. Experiments on real-world datasets show that the proposed model exhibits excellent recommendation performance compared with several baseline models.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.4
• /
• pp.169-178
• /
• 2022

Morphemes are most primitive units in a language that lose their original meaning when segmented into smaller parts. In Korean, a sentence is a sequence of eojeols (words) separated by spaces. Each eojeol comprises one or more morphemes. Korean morphological analysis (KMA) is to divide eojeols in a given Korean sentence into morpheme units. It also includes assigning appropriate part-of-speech(POS) tags to the resulting morphemes. KMA is one of the most important tasks in Korean natural language processing (NLP). Improving the performance of KMA is closely related to increasing performance of Korean NLP tasks. Recent research on KMA has begun to adopt the approach of machine translation (MT) models. MT is to convert a sequence (sentence) of units of one domain into a sequence (sentence) of units of another domain. Neural machine translation (NMT) stands for the approaches of MT that exploit neural network models. From a perspective of MT, KMA is to transform an input sequence of units belonging to the eojeol domain into a sequence of units in the morpheme domain. In this paper, we propose a deep learning model for KMA. The backbone of our model is based on the BERT-fused model which was shown to achieve high performance on NMT. The BERT-fused model utilizes Transformer, a representative model employed by NMT, and BERT which is a language representation model that has enabled a significant advance in NLP. The experimental results show that our model achieves 98.24 F1-Score.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.57-67
• /
• 2024

This research investigates training methods for large language models to accurately identify sentiments and comprehend information about increasing and decreasing fluctuations in the financial domain. The main goal is to identify suitable datasets that enable these models to effectively understand expressions related to financial increases and decreases. For this purpose, we selected sentences from Wall Street Journal that included relevant financial terms and sentences generated by GPT-3.5-turbo-1106 for post-training. We assessed the impact of these datasets on language model performance using Financial PhraseBank, a benchmark dataset for financial sentiment analysis. Our findings demonstrate that post-training FinBERT, a model specialized in finance, outperformed the similarly post-trained BERT, a general domain model. Moreover, post-training with actual financial news proved to be more effective than using generated sentences, though in scenarios requiring higher generalization, models trained on generated sentences performed better. This suggests that aligning the model's domain with the domain of the area intended for improvement and choosing the right dataset are crucial for enhancing a language model's understanding and sentiment prediction accuracy. These results offer a methodology for optimizing language model performance in financial sentiment analysis tasks and suggest future research directions for more nuanced language understanding and sentiment analysis in finance. This research provides valuable insights not only for the financial sector but also for language model training across various domains.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.97-113
• /
• 2016

Due to the growth of internet data and the rapid development of internet technology, "big data" analysis is actively conducted to analyze enormous data for various purposes. Especially in recent years, a number of studies have been performed on the applications of text mining techniques in order to overcome the limitations of existing structured data analysis. Various studies on sentiment analysis, the part of text mining techniques, are actively studied to score opinions based on the distribution of polarity of words in documents. Usually, the sentiment analysis uses sentiment dictionary contains positivity and negativity of vocabularies. As a part of such studies, this study tries to construct sentiment dictionary which is customized to specific data domain. Using a common sentiment dictionary for sentiment analysis without considering data domain characteristic cannot reflect contextual expression only used in the specific data domain. So, we can expect using a modified sentiment dictionary customized to data domain can lead the improvement of sentiment analysis efficiency. Therefore, this study aims to suggest a way to construct customized dictionary to reflect characteristics of data domain. Especially, in this study, movie review data are divided by genre and construct genre-customized dictionaries. The performance of customized dictionary in sentiment analysis is compared with a common sentiment dictionary. In this study, IMDb data are chosen as the subject of analysis, and movie reviews are categorized by genre. Six genres in IMDb, 'action', 'animation', 'comedy', 'drama', 'horror', and 'sci-fi' are selected. Five highest ranking movies and five lowest ranking movies per genre are selected as training data set and two years' movie data from 2012 September 2012 to June 2014 are collected as test data set. Using SO-PMI (Semantic Orientation from Point-wise Mutual Information) technique, we build customized sentiment dictionary per genre and compare prediction accuracy on review rating. As a result of the analysis, the prediction using customized dictionaries improves prediction accuracy. The performance improvement is 2.82% in overall and is statistical significant. Especially, the customized dictionary on 'sci-fi' leads the highest accuracy improvement among six genres. Even though this study shows the usefulness of customized dictionaries in sentiment analysis, further studies are required to generalize the results. In this study, we only consider adjectives as additional terms in customized sentiment dictionary. Other part of text such as verb and adverb can be considered to improve sentiment analysis performance. Also, we need to apply customized sentiment dictionary to other domain such as product reviews.

• The Journal of Society for e-Business Studies
• /
• v.18 no.2
• /
• pp.95-110
• /
• 2013

This paper presents a novel feature engineering technique that can improve the conventional machine learning-based text classification systems. The proposed method extends the initial set of features by using hyperlink relationships in order to effectively categorize hypertext web documents. Web documents are connected to each other through hyperlinks, and in many cases hyperlinks exist among highly related documents. Such hyperlink relationships can be used to enhance the quality of features which consist of classification models. The basic idea of the proposed method is to generate a sort of ed concept feature which consists of a few raw feature words; for this, the method computes the semantic similarity between a target document and its neighbor documents by utilizing hierarchical relationships in the WordNet ontology. In developing classification models, the ed concept features are equated with other raw features, and they can play a great role in developing more accurate classification models. Through the extensive experiments with the Web-KB test collection, we prove that the proposed methods outperform the conventional ones.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.9
• /
• pp.39-47
• /
• 2012

The opinion of user on a certain product is expressed in positive/negative sentiments for specific features of it. In some cases, they are expressed for a holistic part of homogeneous specific features, or expressed for product itself. Therefore, in the area of opinion mining, name of opinion features to be extracted are specific feature names, holonyms for theses specific features, and product names. However, when the opinion target is described with product name or holonym, sometimes it may not match feature name of opinion sentence to true opinion target intended by the reviewer. In this paper, we present a method to extract opinion targets from opinion sentences. Most importantly, we propose a method to extract true target from the feature names mismatched to a intended target. First, we extract candidate opinion pairs using dependency relation between words, and then select feature names frequently mismatched to opinion target. Each selected opinion feature name is replaced to a specific feature intended by the reviewer. Finally, in order to extract relevant opinion features from the whole candidate opinion pairs including modified opinion feature names, candidate opinion pairs are rearranged by the order of user's interest.

• Journal of Internet Computing and Services
• /
• v.19 no.6
• /
• pp.21-29
• /
• 2018

In the era of the 4-th industrial revolution, the concept of smart factory is emerging. There are efforts to predict the occurrences of facility errors which have negative effects on the utilization and productivity by using data analysis. Data composed of the situation of a facility error and the type of the error, called the facility error log, is required for the prediction. However, in many manufacturing companies, the types of facility error are not precisely defined and categorized. The worker who operates the facilities writes the type of facility error in the form with unstructured text based on his or her empirical judgement. That makes it impossible to analyze data. Therefore, this paper proposes a framework for constructing a phrase network to support the identification and classification of facility error types by using facility error logs written by operators. Specifically, phrase indicating the types are extracted from text data by using dictionary which classifies terms by their usage. Then, a phrase network is constructed by calculating the similarity between the extracted phrase. The performance of the proposed method was evaluated by using real-world facility error logs. It is expected that the proposed method will contribute to the accurate identification of error types and to the prediction of facility errors.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.251-273
• /
• 2022

Recently, research on applying text analysis to deep learning has steadily continued. In particular, researches have been actively conducted to understand the meaning of words and perform tasks such as summarization and sentiment classification through a pre-trained language model that learns large datasets. However, existing pre-trained language models show limitations in that they do not understand specific domains well. Therefore, in recent years, the flow of research has shifted toward creating a language model specialized for a particular domain. Domain-specific pre-trained language models allow the model to understand the knowledge of a particular domain better and reveal performance improvements on various tasks in the field. However, domain-specific further pre-training is expensive to acquire corpus data of the target domain. Furthermore, many cases have reported that performance improvement after further pre-training is insignificant in some domains. As such, it is difficult to decide to develop a domain-specific pre-trained language model, while it is not clear whether the performance will be improved dramatically. In this paper, we present a way to proactively check the expected performance improvement by further pre-training in a domain before actually performing further pre-training. Specifically, after selecting three domains, we measured the increase in classification accuracy through further pre-training in each domain. We also developed and presented new indicators to estimate the specificity of the domain based on the normalized frequency of the keywords used in each domain. Finally, we conducted classification using a pre-trained language model and a domain-specific pre-trained language model of three domains. As a result, we confirmed that the higher the domain specificity index, the higher the performance improvement through further pre-training.

• Journal of Intelligence and Information Systems
• /
• v.22 no.3
• /
• pp.45-69
• /
• 2016

Recently, sentiment analysis using open Internet data is actively performed for various purposes. As online Internet communication channels become popular, companies try to capture public sentiment of them from online open information sources. This research is conducted for the purpose of analyzing pulbic sentiment of Korean Top-10 companies using a multi-categorical sentiment lexicon. Whereas existing researches related to public sentiment measurement based on big data approach classify sentiment into dimensions, this research classifies public sentiment into multiple categories. Dimensional sentiment structure has been commonly applied in sentiment analysis of various applications, because it is academically proven, and has a clear advantage of capturing degree of sentiment and interrelation of each dimension. However, the dimensional structure is not effective when measuring public sentiment because human sentiment is too complex to be divided into few dimensions. In addition, special training is needed for ordinary people to express their feeling into dimensional structure. People do not divide their sentiment into dimensions, nor do they need psychological training when they feel. People would not express their feeling in the way of dimensional structure like positive/negative or active/passive; rather they express theirs in the way of categorical sentiment like sadness, rage, happiness and so on. That is, categorial approach of sentiment analysis is more natural than dimensional approach. Accordingly, this research suggests multi-categorical sentiment structure as an alternative way to measure social sentiment from the point of the public. Multi-categorical sentiment structure classifies sentiments following the way that ordinary people do although there are possibility to contain some subjectiveness. In this research, nine categories: 'Sadness', 'Anger', 'Happiness', 'Disgust', 'Surprise', 'Fear', 'Interest', 'Boredom' and 'Pain' are used as multi-categorical sentiment structure. To capture public sentiment of Korean Top-10 companies, Internet news data of the companies are collected over the past 25 months from a representative Korean portal site. Based on the sentiment words extracted from previous researches, we have created a sentiment lexicon, and analyzed the frequency of the words coming up within the news data. The frequency of each sentiment category was calculated as a ratio out of the total sentiment words to make ranks of distributions. Sentiment comparison among top-4 companies, which are 'Samsung', 'Hyundai', 'SK', and 'LG', were separately visualized. As a next step, the research tested hypothesis to prove the usefulness of the multi-categorical sentiment lexicon. It tested how effective categorial sentiment can be used as relative comparison index in cross sectional and time series analysis. To test the effectiveness of the sentiment lexicon as cross sectional comparison index, pair-wise t-test and Duncan test were conducted. Two pairs of companies, 'Samsung' and 'Hanjin', 'SK' and 'Hanjin' were chosen to compare whether each categorical sentiment is significantly different in pair-wise t-test. Since category 'Sadness' has the largest vocabularies, it is chosen to figure out whether the subgroups of the companies are significantly different in Duncan test. It is proved that five sentiment categories of Samsung and Hanjin and four sentiment categories of SK and Hanjin are different significantly. In category 'Sadness', it has been figured out that there were six subgroups that are significantly different. To test the effectiveness of the sentiment lexicon as time series comparison index, 'nut rage' incident of Hanjin is selected as an example case. Term frequency of sentiment words of the month when the incident happened and term frequency of the one month before the event are compared. Sentiment categories was redivided into positive/negative sentiment, and it is tried to figure out whether the event actually has some negative impact on public sentiment of the company. The difference in each category was visualized, moreover the variation of word list of sentiment 'Rage' was shown to be more concrete. As a result, there was huge before-and-after difference of sentiment that ordinary people feel to the company. Both hypotheses have turned out to be statistically significant, and therefore sentiment analysis in business area using multi-categorical sentiment lexicons has persuasive power. This research implies that categorical sentiment analysis can be used as an alternative method to supplement dimensional sentiment analysis when figuring out public sentiment in business environment.