• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.9
• /
• pp.4684-4705
• /
• 2019

Collaborative filtering recommender systems are vulnerable to shilling attacks in which malicious users may inject biased profiles to promote or demote a particular item being recommended. To tackle this problem, many robust collaborative recommendation methods have been presented. Unfortunately, the robustness of most methods is improved at the expense of prediction accuracy. In this paper, we construct a robust Bayesian probabilistic matrix factorization model for collaborative filtering recommender systems by incorporating the detection of user anomaly rating behaviors. We first detect the anomaly rating behaviors of users by the modified K-means algorithm and target item identification method to generate an indicator matrix of attack users. Then we incorporate the indicator matrix of attack users to construct a robust Bayesian probabilistic matrix factorization model and based on which a robust collaborative recommendation algorithm is devised. The experimental results on the MovieLens and Netflix datasets show that our model can significantly improve the robustness and recommendation accuracy compared with three baseline methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.721-722
• /
• 2009

인터넷 시장이 빠르게 성장함에 따라 사용자들의 참여도가 매우 높아졌다. 인터넷 사용자들은 시장의 상품과 더불어 영화와 같은 문화 생활에 관한 의견을 웹 상에 표현하기 시작했고, 현재에 들어 그 양이 방대해 졌다. 본 논문에서는 사용들이 작성한 영화에 관한 리뷰를 모아 방대한 양에서 유용한 정보를 효율적으로 도출하고 요약해서 사용자에게 제공하는 방법을 제안한다.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.379-383
• /
• 2021

감정 분석은 텍스트 데이터에서 인간이 느끼는 감정을 다양한 감정 유형으로 분류하는 것이다. 그러나 많은 연구에서 감정 분석은 긍정과 부정, 또는 중립의 극성을 분류하는 감성 분석의 개념과 혼용되고 있다. 본 연구에서는 텍스트에서 느껴지는 감정들을 다양한 감정 유형으로 분류한 감정 말뭉치를 구축하였는데, 감정 말뭉치를 구축하기 위해 심리학 모델을 기반으로 분류한 감정 어휘 사전을 사용하였다. 9가지 감정 유형으로 분류된 한국어 감정 어휘 사전을 바탕으로 한국어 영화 리뷰 말뭉치에 9가지 감정 유형의 감정을 태깅하여 감정 분석 말뭉치를 구축하고, KcBert에 학습시켰다. 긍정과 부정으로 분류된 데이터로 사전 학습된 KcBert에 9개의 유형으로 분류된 데이터를 학습시켜 기존 모델과 성능 비교를 한 결과, KcBert는 다중 분류 모델에서도 우수한 성능을 보였다.

• Asia Marketing Journal
• /
• v.26 no.1
• /
• pp.23-30
• /
• 2024

Firms often engage in manipulating online reviews as a promotional activity to influence consumers' evaluation on their products. With the prevalence of the promotional activities, consumers may notice and discount the reviews generated by the promotional activities. Discounting the firm-generating reviews may cause systematic measurement errors in the valence variable and lead to a negative bias when estimating the effect of consumers' organic reviews on demand. To correct the bias, this study proposes including product-specific bias-correction terms representing the proportion of extreme reviews in analysis. For illustration, the proposed method is applied to a demand model for data of movies released in South Korea. The results confirm a negative bias in the estimate of the valence sensitivity of demand. The negative bias potentially leads to an underestimation of the magnitude of the contagion effect through social interactions, a key component of evaluating the value of a satisfied consumer.

• Cartoon and Animation Studies
• /
• s.42
• /
• pp.125-150
• /
• 2016

In the 2000s, Korean cinema maintained the industrial growth on the strength of 10 million audiences and the successful Korean blockbuster. A variety of film materials such as history, politics, social issues, and the division between North and South Korea ideology were addressed in the movies, and one of the movies based on a true story was popular. Accordingly, external size of the film industry has been gradually expanding. Korean public movies have a firm position in the film market, and a diversity of discussion is made as the movies spontaneously get attention. Also, the influence of social media and media which recognized far-reaching powers of public films causes a political and social change, but it also provokes ideology controversy. Objectivity of this study is first to discuss factors and initiating causes that Korean movies have firmly settled as public movies since late 1990s. Secondly, this study considers relation among media, social media. and Korean public film that aroused more arguments on politics, society, and history in 2000s. The powerful influence of films on society not only leads political change but also affects awareness change of audience and the role of social media.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.15-30
• /
• 2023

Recently, generative models based on the Transformer architecture, such as ChatGPT, have been gaining significant attention. The Transformer architecture has been applied to various neural network models, including Google's BERT(Bidirectional Encoder Representations from Transformers) sentence generation model. In this paper, a method is proposed to create a text binary classification model for determining whether a comment on Korean movie review is positive or negative. To accomplish this, a pre-trained multilingual BERT sentence generation model is fine-tuned and transfer learned using a new Korean training dataset. To achieve this, a pre-trained BERT-Base model for multilingual sentence generation with 104 languages, 12 layers, 768 hidden, 12 attention heads, and 110M parameters is used. To change the pre-trained BERT-Base model into a text classification model, the input and output layers were fine-tuned, resulting in the creation of a new model with 178 million parameters. Using the fine-tuned model, with a maximum word count of 128, a batch size of 16, and 5 epochs, transfer learning is conducted with 10,000 training data and 5,000 testing data. A text sentiment binary classification model for Korean movie review with an accuracy of 0.9582, a loss of 0.1177, and an F1 score of 0.81 has been created. As a result of performing transfer learning with a dataset five times larger, a model with an accuracy of 0.9562, a loss of 0.1202, and an F1 score of 0.86 has been generated.

• The Journal of the Korea Contents Association
• /
• v.18 no.10
• /
• pp.78-98
• /
• 2018

On the 855 pure Korean commercial fictional movies, excluding diversity films, released in Korea from 2004 to August 2017, I conducted deciles distribution analysis of box office performance of those movies and average box office performance of directors, producers and lead actors who involved in making them. Deciles distribution analysis of average box office performance might be helpful to predict their next box office performance of newly produced Korean movies and to evaluate their contribution to box office performance. In baseball, the various index such as winning rate, on-base percentage, slugging percentage, stolen base percentage, battling average, earned run average is used for predicting and reviewing of professional players. In this study, I evaluate the script's narrative quality by the indirect method of insight and judgment of creative manpower involved in making the movies. For the more productive prediction, direct statistical analysis method on the narrative of the script needs to develop. Time series analysis is required to evaluate the rise and fall of creative manpower and network analysis is also necessary to see the interaction among creative people.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.144-151
• /
• 2020

In a mobile environment, communication takes place via SMS text messages. Vocabularies used in SMS texts can be expected to use vocabularies of different classes from those used in general Korean literary style sentence. For example, in the case of a typical literary style, the sentence is correctly initiated or terminated and the sentence is well constructed, while SMS text corpus often replaces the component with an omission and a brief representation. To analyze these vocabulary usage characteristics, the existing colloquial style corpus and the literary style corpus are used. The experiment compares and analyzes the vocabulary use characteristics of the colloquial corpus SMS text corpus and the Naver Sentiment Movie Corpus, and the written Korean written corpus. For the comparison and analysis of vocabulary for each corpus, the part of speech tag adjective (VA) was used as a standard, and a distinctive collexeme analysis method was used to measure collostructural strength. As a result, it was confirmed that adjectives related to emotional expression such as'good-','sorry-', and'joy-' were preferred in the SMS text corpus, while adjectives related to evaluation expressions were preferred in the Naver Sentiment Movie Corpus. The word embedding was used to automatically construct a sentiment lexicon based on the extracted adjectives with high collostructural strength, and a total of 343,603 sentiment representations were automatically built.

• Journal of Korea Entertainment Industry Association
• /
• v.13 no.8
• /
• pp.181-187
• /
• 2019

The theme of "youth love film" is a strong point in the Korean film industry and has made remarkable achievements in the overseas market. Jae-young Kwak is a famous Korean film director and screenwriter. He is good at shooting "youth love film" with beautiful pictures, exquisite emotions and rich imagination. His films have unique charm and characteristics. This paper expounds the image style, characterization and theme. First of all, from the perspective of image style for this paper, analysis of the film is how to use different narratives and lens language to create a movie's atmosphere, makes the transfer of drunk the oriental beauty. Secondly, the main focus is on the director to find a new way to shape the characters in the film, and pay attention to the description of details, foreshadowing and explaining the plot. Thirdly, based on the theme, this paper analyzes the profound meaning behind romantic films, which reflect the yearning and pursuit of modern people, and people should pay attention to and think deeply. In the modern economic globalization, countries culture mutual exchanges and cooperation, South Korea many outstanding movie worth exploring and research, by studying the Jae-young director Kwak youth love films, excavate its popular film in the market of reason, analysis of the works reflect the aesthetic characteristics, which sums up the experience of the youth love film.

• Journal of the Korean Society of Costume
• /
• v.31
• /
• pp.145-164
• /
• 1997

Cinema costume can stimulate the public to synchronize with and replicate what they see in the movies. Especially when the public thinks the stars in the movies . Especially when the public thinks the stars in the movies are beautiful or dramatic the styles of cinema costumes powerfully affect them. Thus cinema costume designers quickey become fashion designers. Edith Head who strongly believed cinema costumes to be an effective means of portraying the Character's personality and psychological make-up was an innovative designer whose costumes manifested this belief. She enhanced the character's individual image and distinction for the public through her artistic design in costumes. She adapted a realistic approach but her costumes invariably produced aesthetic ef-fects. Since the Academy Awards added the category of costume design she led the field with 35 nominations and eight oscars during 33 years of her productive work. Edith head was a dominant figure in costume design; her creations were highly renowned as an art form while they gained popularity. The purpose of this study was to reveal the ef-fects of Edith Head's cinema costume designs on fashion. This study was also to emphasize the powerful influence on ashion of cinema costume and cinema costume designers in general. The study was done through literature movies for which she created costumes fashion magazines and Edith Head's original sketches. The influence of Edith Head's costume de-sign on fashion is as follows: 1. She introduced and lead the casual fashion in the 1950's combining simplicity and sophistication. 2. She altered uniforms and reconstructed public clothing for multiple functions. 3. She rearranged high school prom dresses through the movie 'A Place in the Sun" to en-hance characteristic of each gender and the Dandy Look of the 1930's into men's apparel in the 1970's through the movie "The Sting" 4. She expanded the traditional style in the 1940's and became a leader in ecological style. 5. She adapted glitter from astronaut's ap-parel into public clothing. 6. She resturctured Sabrina Pants in the 1960's to accentuate the individuals strength by overcoming one's weakness without minimi-zing it's aesthetic appeal and reintroduced the pants in the 1990's. 7. She introduced elegant negligees to gen-eral public and maintained it in the public do-main. As it is revealed Edith Head's costume de-sign not only impacted the field of fashion dur-ing her time but also is presently influencing modern fashion throughout the world as it is witnessed by repeated reintroduction of her fashion styles. Reviewing her cinema costumes it is concluded that cinema costumes it is concluded that cinema costumes powerfully af-fect the public through a visual channel more than any other routes. Cinema costumes can become the origin of fashion by stimulating the public to synchronize with and replicate what they see in the movies. Since cinema costumes can launch fashion trend it requires further research. Based on what is revealed in this study. it would be beneficial to examine how cinema costumes affect people socially and culturally and how they could provide resources for re-search in fashion trends. It is also the writers opinion that there should be more designers such as Edith head who could lead the field of costume design into the twenty-first century.