Negative Keywords Research Articles

인천 관광지의 속성별 평가를 위한 딥러닝 기반 감성분석

This study applied the IPA(importance-performance analysis) methodology to unstructured text data to evaluate tourists' satisfaction levels with Incheon attractions. Focusing on attractions within the maintenance and improvement-needed areas identified by IPA, it extracted key positive and negative keywords and sentences directly from tourists' opinions. By applying a text-based approach to evolving IPA research, this exploratory study assessed the concrete satisfaction and dissatisfaction elements felt by tourists. The study limited its scope to 15 Incheon attractions and analyzed 114,007 sentences from Naver blogs written by tourists. Satisfaction levels were derived through deep learning sentiment analysis of the text, while the importance of each attraction was determined by the TF-IDF(term frequency-inverse document frequency) ratio of the attraction names in the text categorized by attributes such as attraction, transportation, activities, and food/drink. Key positive and negative keywords and sentences were extracted using the TextRank algorithm, focusing on attractions in the maintenance and improvementneeded areas. The findings reveal that experiential elements of attractions act as differentiating factors, while their absence can provoke negative reactions from tourists. Moreover, content reflecting local culture and history, the development of family-friendly experiential attractions, parking space expansion, and food service improvements emerged as crucial elements for attraction development.

Normalisation of evacuation under slow emergencies: The imposed story of ‘Beautiful New Hong Kong’

AbstractHong Kong, in contrast to its previous image as a glamourous global city, has recently been associated with negative keywords such as oppression, fear, violence and even human rights emergency, following the 2019 Anti‐Extradition Law Movement and later the implementation of National Security Law (NSL) in 2020. Meanwhile, the Hong Kong Special Area Administration presented a post‐NSL policy direction as promising a ‘Beautiful New Hong Kong’. This paper is aimed at understanding how Hongkongers have continuously lived with/against the imposed (re)writing of Hong Kong. We examined how Hongkongers have been taking initiatives to raise awareness about the ‘slow emergencies’ in Hong Kong and to counter it in various forms of ‘evacuation’, whether they are on the move or staying put in place, in order to pursue future‐making. We carried out a multi‐sited study on the two kinds of Hongkongers between January 2020 and August 2023. We talked to those who already left Hong Kong for Taiwan, the UK, Canada, and so forth, and with those who were debating about relocation and at the same time preparing for departure if necessary. We strategically read ‘evacuation’ in two senses: First, evacuation responds to emergencies and therefore by adopting ‘evacuation’ is itself a disagreement with the Beautiful New Hong Kong policy as curated by the state. Second, evacuation responds to geography of future and politics of simultaneity. We conceptualised ‘normalisation of evacuation’ to understand the future‐making behind the move in Hong Kong. The particular kind of evacuation discussed took shape in two forms, relocation elsewhere and reorganisation in situ, both of which, in our analysis, are demonstrating Hongkongers' agency in pursuing geographies of future beyond the state‐led agenda.

Improving Persian Opinion Mining based on Polarity and Balancing Positive and Negative Keywords (case study: Digikala reviews for mobile)

Improving Persian Opinion Mining based on Polarity and Balancing Positive and Negative Keywords (case study: Digikala reviews for mobile)

Data-driven budget allocation of retail media by ad product, funnel metric, and brand size

Sellers on online marketplaces such as Amazon.com use a variety of retail and retail media advertising services to improve their brand performance, including awareness, consideration, and revenue. But how can they measure their progress and drive these metrics? For 122,000 brands, we measure Amazon shoppers’ brand awareness, consideration, and purchases and test how they change with ad and retail actions. Furthermore, we compare these brands’ past media mix with the recommended allocation based on the model’s coefficients. We find that new product launches and upper-funnel retail media advertising are particularly effective for small brands. Medium-sized and large brands benefit most from lower-funnel advertising. For the funnel stages, all three metrics benefit from the number of new reviews, % discount, negative keywords, and geo-reach campaigns. These results are robust across different product categories, but we find notable differences in how upper- and middle-funnel ad products succeed in driving sales.

Temporal and Spatial Analysis of Negative Emotions in China during the COVID-19 Pandemic.

This research aims to explore the spatiotemporal distribution patterns of negative emotions in mainland China during different stages of the COVID-19 pandemic and the external factors influencing this clustering. Using Baidu Index data for 91 negative emotion keywords, a retrospective geographic analysis was conducted across Chinese provinces from 14 October 2019 to 7 July 2022. Four spatial analysis methods (Global Moran's Index, Local Moran's Index, Bivariate Global Moran's Index, and Bivariate Local Moran's Index) are employed to identify potential clustering patterns and influencing factors of negative emotions at different stages. The results indicate that the COVID-19 pandemic significantly intensified the clustering effect of negative emotions in China, particularly with a more pronounced radiation effect in northwestern provinces. Spatial positive correlations are observed between pandemic-related Baidu indices (pandemic Baidu index, government Baidu index, nucleic acid Baidu index) and negative emotions. These findings contribute to understanding the spatiotemporal distribution characteristics of negative emotions in China post the COVID-19 outbreak and can guide the allocation of psychological resources during emergencies, thereby promoting social stability.

STOCK MARKET ANALYSIS

The abstract aims to address the correlation between stock market movements and public sentiments expressed on Twitter. It delves into the utilization of sentiment analysis and supervised machine learning techniques to explore this connection. The study leverages Word2vec for textual representation, examining how shifts in stock prices align with sentiments expressed in tweets about specific companies. The investigation underscores the potential impact of positive news and social media sentiments on stock prices, emphasizing a demonstrated correlation between fluctuations in stock prices and sentiments conveyed in Twitter.- Keywords: Hashtag Collection, Data Collection, Real-Time Stock History Data, Positive Keywords, Negative Keywords, Polarity Computation, Sentiment Analysis , Sentiment Index Computation, Sentiment Discrepancy Index, Price Prediction, Yahoo Finance API

Data-mining analysis of media frame effects on social perception of schizophrenia renaming in Korea

BackgroundIn 2011, Korean Neuropsychiatric Association renamed schizophrenia from ‘mind split disorder’ (‘Jungshinbunyeolbyung’ in Korean) to ‘attunement disorder’ (‘Johyeonbyung’ in Korean), in a strategic way to reduce social stigma toward people with schizophrenia. However, there remains an elusive consensus that how the renaming effort has contributed to changes in the social perception of schizophrenia in Korea.MethodsWith this regard, we explored whether media frames alter the social perception, in ways of respecting or disrespecting schizophrenia patients before and after the renaming. This study extensively investigated media keywords related to schizophrenia across the time by applying both language and epidemiologic analyses.ResultsIn results, the media keywords have been negatively described for schizophrenia patients both before and after the renaming. Further, from an analysis using the regression model, a significant correlation was observed between the frequency of negative keywords and the hospitalization frequency of schizophrenia patients.ConclusionsThese findings suggest that the social perception of schizophrenia has been scarcely changed, but rather remained negatively biased against schizophrenia patients, in spite of the renaming effort. Notably, the biased media frames have been demonstrated to negatively impact on the social perception, and even on the medical use patterns of general schizophrenia patients. In conclusion, we suggest that the unbiased media frames along with the renaming effort may collectively help reduce the negative social perception of schizophrenia.Trial registrationThis study was approved from the Institute of Review Board (IRB) of the Yoing-In Mental Hospital (IRB No. YIMH-IRB-2019-02).

The Analysis Actual Use and Perception of Class Using Coding of Chemistry Teacher

This study investigated the actual use and perceptions of class using coding for chemistry teachers. To this end, a questionnaire survey consisting of a total of 7 questions was conducted targeting 45 teachers. The research results were as follows. First, chemistry teachers had the most cases of using digital devices with ‘Utilizing PPT’, and the rate of experiencing coding-related education was not high during the teacher period as well as the pre-teacher period. As a result of examining the level of use, only two teachers were actually using coding in class. Second, chemistry teachers recognized that classes using coding would be effective for ‘increasing interest of class’ and ‘offering a variety of learning experiences’. As a result of open-ended response analysis of classes using coding, the frequency of negative keywords was higher than that of positive keywords, and the instructor's low readiness for coding was the biggest barrier to implementing classes using coding. Through these results, teachers recognized the value of classes using coding as a new learning tool, but in terms of actual use, teachers' preparation or conditions have not yet been formed, so the practice of classes using coding is low. Based on this, we discussed ways to activate classes using coding.

FinChain-BERT: A High-Accuracy Automatic Fraud Detection Model Based on NLP Methods for Financial Scenarios

This research primarily explores the application of Natural Language Processing (NLP) technology in precision financial fraud detection, with a particular focus on the implementation and optimization of the FinChain-BERT model. Firstly, the FinChain-BERT model has been successfully employed for financial fraud detection tasks, improving the capability of handling complex financial text information through deep learning techniques. Secondly, novel attempts have been made in the selection of loss functions, with a comparison conducted between negative log-likelihood function and Keywords Loss Function. The results indicated that the Keywords Loss Function outperforms the negative log-likelihood function when applied to the FinChain-BERT model. Experimental results validated the efficacy of the FinChain-BERT model and its optimization measures. Whether in the selection of loss functions or the application of lightweight technology, the FinChain-BERT model demonstrated superior performance. The utilization of Keywords Loss Function resulted in a model achieving 0.97 in terms of accuracy, recall, and precision. Simultaneously, the model size was successfully reduced to 43 MB through the application of integer distillation technology, which holds significant importance for environments with limited computational resources. In conclusion, this research makes a crucial contribution to the application of NLP in financial fraud detection and provides a useful reference for future studies.

Automated sentimental analysis using heuristic-based CNN-BiLSTM for E-commerce dataset

The sentimental analysis is processed according to structural detection, extraction, quantification, and evaluation of impacts and information present in natural language processing, biometric data, computer-language society, and text analysis. Advanced implementations in data analytics occur, which allows for identifying the underlining trends based on certain efficient computational models. Some of the social and e-commerce platforms consist of numerous reviews regarding the online products that are posted by the end-users. This helps the developers to understand priceless insight during the process of product design. The most significant aim of this work is to develop a Heuristic-based sentimental analysis for E-Commerce data. Here, the pre-processing of input E-Commerce data is performed by blank space removal, stop word removal, and stemming. Here, the statistical features are extracted using Cross Similarity Score (CSS) and Joint Similarity Score (JSS) regarding “positive, negative, and neutral keywords”. In addition, the word2vec features are also extracted. From the two sets of features, “Convolutional Neural Network (CNN) with Bidirectional Long Short Term Memory (BiLSTM)” is separately used that is named as Optimal Hybrid CNN-BiLSTM (OH-CNN-BiLSTM), and performed the hybridization of two models for the sentimental analysis. The designed CNN-BiLSTM technique is ensured better outcomes when processing long text as it provides advantages through the ability of CNN to extract the features and superiority of Bi-LSTM in learning the bidirectional dependencies of the long-term text. The main novelty of this work is to develop an improved Galactic Swarm Optimization (IGSO) algorithm for improving the hybridized model, as it provides faster convergence when getting the accurate solution on a wide range of dimensional space and is also able to solve the optimization issues with multimodal benchmark data. The evaluation of the suggested algorithm is done with various E-Commerce data and secures high accuracy on sentimental analysis and prediction while comparing with the traditional models.

태권도 겨루기 심판판정에 대한 지도자의 인식 : 텍스트 마이닝의 적용

The purpose of this study is to analyze the perceptions of Leaders on Taekwondo Kyorugi Judge’s decision. To achieve the direction of this study, 121 leaders registered at 2022 in Korea Taekwondo Association were surveyed. Perceptions of Leaders on Taekwondo Kyorugi Judge’s decision was conducted through an open questionnaire. The survey is composed of six questions. The current study collected 208 keywords, 345 keywords, 280 keywords, 220 keywords, and 351 keywords for each question. In detail, this study. text-mining method was applied in total 1,404 keywords calculation, there were used in research target data. Fairness and confidence level were analyzed through technical statistics to the general characteristics of the research subject. For data processing, technical statistics, frequency and matrix of keywords using SPSS, KrKwic, Neminer 4.0 program, degree centrality was conducted. The conclusions of this study are as follow. improved the quality of the referee’s decision on the score. Second, all part that people need to judge, negative keywords were producted such as ‘Inconsistent’, ‘Objective’, ‘Unreliable’, ‘Unfair’, ‘Game rules’, ‘Standard ambiguity’. Third, ‘Game rules’, ‘Standard ambiguity’, ‘Deducting points’, ‘Video reading’, ‘Fist scoring’, ‘Facial scoring’, ‘Clinch situation’, ‘Inconsistent’, ‘Unfair’, ‘The effect of victory or defeat’, as a result, these keywords have a negative connection relationship with each other. Fourth, fairness and confidence level of judge’s decision was 48.22%. The conclusion of this study can be used as basic data for various.

소셜미디어에 나타난 풍력발전시설의 경관 인식 연구

This study aims to assess the current understanding of the landscape of wind power facilities as renewable energy sources that supply sightseeing, tourism, and other opportunities. Therefore, social media data related to the landscape of wind power facilities experienced by visitors from different regions was analyzed. The analysis results showed that the common characteristics of the landscape of wind power facilities are based on the scale of wind power facilities, the distance between overlook points of wind power facilities, the visual openness of the wind power facilities from the overlook points, and the terrain where the wind power facilities are located. In addition, the preference for wind power facilities is higher in places where the shape of wind power facilities and the surrounding landscape can be clearly seen- flat ground or the sea are considered better landscapes. Negative keywords about the landscape appear on Gade Mountain in Taibai, Meifeng Mountain in Taibai, Taiqi Mountain, and Gyeongju Wind Power Generation Facilities on Gyeongshang Road in Gangwon. The keyword ‘negation’ occurs when looking at wind power facilities at close range. Because of the high angle of the view, viewers can feel overwhelmed seeing the size of the facility and the ridge simultaneously, feeling psychological pressure. On the contrary, positive landscape adjectives are obtained from wind power facilities on flat ground or the sea. Visitors think that the visual volume of the landscape is fully ensured on flat ground or the sea, and it is a symbolic element that can represent the site. This study analyzes landscape awareness based on the opinions of visitors who have experienced wind power facilities. However, wind power facilities are built in different areas. Therefore, landscape characteristics are different, and there are many variables, such as viewpoints and observers, so the research results are difficult to popularize and have limitations. In recent years, landscape damage due to the construction of wind power facilities has become a hot issue, and the domestic methods of landscape evaluation of wind power facilities are unsatisfactory. Therefore, when evaluating the landscape of wind power facilities, the scale of wind power facilities, the inherent natural characteristics of the area where wind power facilities are set up, and the distance between wind power facilities and overlook points are important elements to consider. In addition, wind power facilities are set in the natural environment, which needs to be protected. Therefore, from the landscape perspective, it is necessary to study the landscape of wind power facilities and the surrounding environment.

Analysis of Food Delivery Using Big Data: Comparative Study before and after COVID-19.

This study examined consumers’ change in perception related to food delivery using big data before and after the COVID-19 crisis. This study identified words closely associated with the keyword “food delivery” based on big data from social media and investigated consumers’ perceptions of and needs for food delivery and related issues before and after COVID-19. Results were derived through analysis methods such as text mining analysis, Concor analysis, and sentiment analysis. The research findings can be summarized as follows: In 2019, frequently appearing dining-related words were “dining-out,” “delivery,” “famous restaurant,” “delivery food,” “foundation,” “dish,” “family order,” and “delicious.” In 2021, these words were “delivery,” “delivery food,” “famous restaurant,” “foundation,” “COVID-19,” “dish,” “order,” “application,” and “family.” The analysis results for the food delivery sentimental network based on 2019 data revealed discourses revolving around delicious, delivery food, lunch box, and Korean food. For the 2021 data, discourses revolved around delivery food, recommend, and delicious. The emotional analysis, which extracted positive and negative words from the “food delivery” search word data, demonstrated that the number of positive keywords decreased by 2.85%, while negative keywords increased at the same rate. In addition, compared to the pre-COVID-19 pandemic era, a weakening trend in positive emotions and an increasing trend in negative emotions were detected after the outbreak of the COVID-19 pandemic; sub-emotions under the positive category (e.g., good feelings, joy, interest) decreased in 2021 compared to 2019, whereas sub-emotions under the negative category (e.g., sadness, fear, pain) increased.

A Study of Consumers’ Perceptions of Take-Out Food before and after the COVID-19 Outbreak: Applying Big Data Analysis

This study explored changes in consumers’ perceptions of take-out food before and after the onset of the COVID-19 pandemic using big data collected from social media. Using “take-out food” as a keyword, 18,544 search results were found in 2019, before the COVID-19 outbreak, compared to 20,718 search results in 2021. These keywords were analyzed using text mining, semantic network analysis, CONCOR analysis, and sentiment analysis, respectively, to understand consumers’ perception of take-out food. Using text mining, in 2019, “dining-out” was the most frequent search term associated with take-out food, followed by packing, famous restaurant, family, delicious, menu, and available. In 2021, “dining-out” was again the most popular keyword, followed by packing, famous restaurant, delivery, family, delicious, available, and Corona. A semantic network analysis showed that, in 2019, four categories emerged: delicious, meat, satisfaction, and lunchbox. The same analysis showed that, in 2021, the categories were delicious, meat, good, and home meal. These findings suggest that, after COVID-19, take-out food began to be recognized as a daily meal that can replace home-cooked meals. According to the sentiment analysis, the number of positive keywords decreased by 4.03% after COVID-19, while the number of negative keywords increased at the same rate; regarding the increase in negative keywords, such as sadness, disgust, and fear, since the emergence of COVID-19, consumers’ anxiety about eating out due to the virus was observed. This study can be useful by providing core data and an analysis method necessary for food service companies’ business activities and decision making related to take-out amid consumers’ rapidly changing needs for the dining-out environment caused by COVID-19.

Influence of COVID-19 on student campus ratings: a sentiment analysis

PurposeGiven the disruption of the COIVD-19 pandemic in higher education, this study seeks to understand possible changes in students’ ratings and textual reviews of higher education institutions posted on Niche College Rankings (niche.com) prior to and after the COVID-19 pandemic.Design/methodology/approachThis study utilized a text analytics technique to identify the positive and negative keywords of students’ sentiments expressed in their textual reviews provided on niche.com. After identifying the positive and negative sentimental keywords, this study performed ordinal logistic regressions and analyzed the statistical effects of these positive and negative sentimental keywords on the types of student ratings of a higher education institution.FindingsResults from 15,666 online reviews provided by students on niche.com indicate the following. First, eight positive sentimental keywords such as “outstanding” and “love” have a significant impact on students’ positive ratings of a higher education institution prior to COVID-19, whereas eight positive sentimental keywords such as “amazing” and helpful” have a significant impact on students’ positive ratings of a higher education institution after COVID-19. Second, twenty-eight negative sentimental keywords such as “difficult” and “frustrating” have a significant impact on students’ negative ratings of a higher education institution prior to COVID-19, whereas thirty negative sentimental keywords such as “complex” and “hate” have a significant impact on student negative ratings of a higher education institution after COVID-19.Originality/valueThis study is one of the first few studies investigating higher education institution ratings and reviews provided by students. Additionally, this study provides an understanding of student positive and negative sentiments expressed in textual reviews posted prior to and after the COVID-19 pandemic. By doing so, this study provides a basis for future research seeking to understand student textual reviews of higher education institutions. Additionally, this study offers higher education administrators some recommendations that may foster student positive campus experience while minimizing negative sentiments.

Strategic Crisis Response of Shipping Industry in the Post COVID-19 Era: A Case of the Top 10 Shipping Lines

Shipping has played a pivotal role during the epidemic, ensuring that the global logistics functions without disruption. COVID-19 hit various industries around the world, and shipping was no exception. How the shipping industry responds to the crisis and simultaneously shoulders its respective responsibility in the world’s battling the crisis is thus worth exploring in depth. This study takes the top 10 global container shipping capacity liners, which account for 84.7% of the worldwide capacity, as research objects. A corpus for text analysis was constructed collecting press releases and advisories issued on official websites of these 10 container shipping companies from January 2020 to July 2021. Comparison studies were made horizontally among ten shipping companies and longitudinally for crisis evolution patterns into three sub-corpora of Pre/early-Crisis, Crisis-in-Progression, and Post-COVID-19-Era. Quantitative findings were explored and elaborated further under a comprehensive theoretical framework integrating crisis management and communication, corporate social responsibility (CSR), and maritime management (MM). The extracted positive and negative keywords revealed textual characteristics and emergency response strategies on the part of shipping lines in the Pre/early-Crisis, Crisis-in-Progression, and Post-COVID-19-Era. The inclusion of the themes of pursuing sustainability in the shipping lines’ responses to such worldwide crisis as COVID-19 is out of the common knowledge of crisis management but reveals the commitment and strategies on the part of the industry. The findings provide a reasonably comprehensive picture of the efforts made by large container shipping companies to respond to COVID-19 and the measures taken to soothe stakeholders. This paper extends and relearns crisis management, CSR, and MM theories through integrating the fulfilling of cooperate social responsibilities in maritime management as the cooperate crisis responses, thus proposing the integrity of the three topics. Moreover, management recommendations are provided for shipping company management, IMO, and port authorities.

빅데이터를 활용한 e스포츠 인식 및 요구분석

The purpose of this study is to analyze e-Sports keywords utilizing the big data analytical technique, and provide the basic data required for establishing e-Sports policies. To achieve this purpose of research, text mining, sentiment analysis, and CONCOR analysis were performed. The search volume analysis results for e-Sports keywords showed the maximum search volume as of October 2021. This period is the time of the 2021 LoL World Championships, and the success of the events has been shown to affect e-Sports interest. For keyword frequency analysis results, sports, games, Korea, competitions, athletes, league of legends, venues, battlegrounds, zenji, competitions were shown the highest frequency and derived as key keywords. As a result of sentiment analysis, the positive rate (75.81%) was found to be very high, and that was classified as good, recommendation, new, growth, best, wanted, etc. The negative rate was 24.19%, and negative keywords such as difficult, esoteric, caution, bad, etc. were classified. As a result of the CONCOR analysis, ‘National e-Sports League’, ‘e-Sports Professional Training’, ‘Asian Games’ and ‘Amateur e-Sports Events’ were formed as a group. These results are believed to reflect interests of demand for national and international e-Sports events and fostering the specialized workforce through the professional training courses.

Divorce Fact Detection Based on Internet User Behavior Using Hybrid Systems with Combination of Apriori Algorithm and K-Means Method

An ideal couple should sustain the family's ark till the end of their life without divorce. This study aims at seeking the association between divorce and internet behavior of searching negative keywords. The study observes four keywords, which are porn, sex, gay, and lesbian. We collected keyword usage data from google trend reports and obtained divorce court figures from the Religion Court of Padang. We used the apriori algorithm to reveal the association between divorce and internet behavior observing individual keyword searches and in groups. We used the K-Means algorithm in classifying negative word searches and divorce trial numbers from a group of existing data. We also investigate the combination of the apriori algorithm and the K-Means method to detect divorce facts and the behavior of internet users. The combined method has been successful in revealing the positive association between divorce facts and the behavior of internet users.

PESTICIDE EXPOSURE AND HEALTH PROBLEMS AMONG WOMEN IN REPRODUCTIVE AGE: A LITERATURE REVIEW

Pesticides are persistent organic pollutants that are resistant to environmental degradation and have a negative impact on human health and the environment. Exposure to chemicals from pesticides, can be accepted by anyone, including women, both women of reproductive age and women who are pregnant. This study was a literature review study with data collection process was carried out by obtaining articles relevant to the research objectives in the ScienceDirect databases. The keywords used include “Pesticide for Women” and “Pesticide for Women Farmer”. After that, 10 articles that match the research objectives were selected. Exposure to pesticides during early pregnancy may increase the risk of adverse birth outcomes such as preterm delivery, small for gestational age, and low birth weight. Pesticide use is associated with a wide range of risks to human health and the environment due to poor handling practices, inadequate knowledge about pesticide toxicity, and the widespread availability of banned or unauthorized chemicals, and that impacts are also largely underestimated and underreported. Multiple approaches including implementation of the regulation, choosing alternative biopesticides, using personal protective equipment, and regular counseling on all aspects of pesticides, their impact on health are recommended to minimize pesticide's negative effects   Keywords: pesticide, women, reproductive

Lexicon Based Sentiment Analysis in Indonesia Languages : A Systematic Literature Review

This systematic literature review aims to determine the trend of lexicon based sentiment analysis research in Indonesian Language in the last two years. The focus of the study is on the understanding of preprocessing used in lexicon-based sentiment analysis studies in the last two years, the lexicon used in these studies, and classification accuracy. The main question in this SLR : what techniques of lexicon based sentiment analysis will provide the highest accuracy. The most widely used preprocessing methods in previous research are tokenization, case conversion, stemming, remove punctuation, remove stop word, remove or replace emoji and emoticons, and normalization or slangword conversion. The sentiment labeling process in previous studies calculated based on the comparison of the number of negative sentiment keywords with positive sentiment keywords in one sentence. The maximum accuracy from previous study is 90%. The most widely used lexicon is NRC and Inset which is a lexicon dictionary in Indonesian. Knowledge of this can be used to propose a better model for lexicon based sentiment analysis in Indonesian Languages.