Sentiment Analysis of Movie Review Using Integrated CNN-LSTM Mode (CNN-LSTM 조합모델을 이용한 영화리뷰 감성분석)
-
- Journal of Intelligence and Information Systems
- /
- v.25 no.4
- /
- pp.141-154
- /
- 2019
Rapid growth of internet technology and social media is progressing. Data mining technology has evolved to enable unstructured document representations in a variety of applications. Sentiment analysis is an important technology that can distinguish poor or high-quality content through text data of products, and it has proliferated during text mining. Sentiment analysis mainly analyzes people's opinions in text data by assigning predefined data categories as positive and negative. This has been studied in various directions in terms of accuracy from simple rule-based to dictionary-based approaches using predefined labels. In fact, sentiment analysis is one of the most active researches in natural language processing and is widely studied in text mining. When real online reviews aren't available for others, it's not only easy to openly collect information, but it also affects your business. In marketing, real-world information from customers is gathered on websites, not surveys. Depending on whether the website's posts are positive or negative, the customer response is reflected in the sales and tries to identify the information. However, many reviews on a website are not always good, and difficult to identify. The earlier studies in this research area used the reviews data of the Amazon.com shopping mal, but the research data used in the recent studies uses the data for stock market trends, blogs, news articles, weather forecasts, IMDB, and facebook etc. However, the lack of accuracy is recognized because sentiment calculations are changed according to the subject, paragraph, sentiment lexicon direction, and sentence strength. This study aims to classify the polarity analysis of sentiment analysis into positive and negative categories and increase the prediction accuracy of the polarity analysis using the pretrained IMDB review data set. First, the text classification algorithm related to sentiment analysis adopts the popular machine learning algorithms such as NB (naive bayes), SVM (support vector machines), XGboost, RF (random forests), and Gradient Boost as comparative models. Second, deep learning has demonstrated discriminative features that can extract complex features of data. Representative algorithms are CNN (convolution neural networks), RNN (recurrent neural networks), LSTM (long-short term memory). CNN can be used similarly to BoW when processing a sentence in vector format, but does not consider sequential data attributes. RNN can handle well in order because it takes into account the time information of the data, but there is a long-term dependency on memory. To solve the problem of long-term dependence, LSTM is used. For the comparison, CNN and LSTM were chosen as simple deep learning models. In addition to classical machine learning algorithms, CNN, LSTM, and the integrated models were analyzed. Although there are many parameters for the algorithms, we examined the relationship between numerical value and precision to find the optimal combination. And, we tried to figure out how the models work well for sentiment analysis and how these models work. This study proposes integrated CNN and LSTM algorithms to extract the positive and negative features of text analysis. The reasons for mixing these two algorithms are as follows. CNN can extract features for the classification automatically by applying convolution layer and massively parallel processing. LSTM is not capable of highly parallel processing. Like faucets, the LSTM has input, output, and forget gates that can be moved and controlled at a desired time. These gates have the advantage of placing memory blocks on hidden nodes. The memory block of the LSTM may not store all the data, but it can solve the CNN's long-term dependency problem. Furthermore, when LSTM is used in CNN's pooling layer, it has an end-to-end structure, so that spatial and temporal features can be designed simultaneously. In combination with CNN-LSTM, 90.33% accuracy was measured. This is slower than CNN, but faster than LSTM. The presented model was more accurate than other models. In addition, each word embedding layer can be improved when training the kernel step by step. CNN-LSTM can improve the weakness of each model, and there is an advantage of improving the learning by layer using the end-to-end structure of LSTM. Based on these reasons, this study tries to enhance the classification accuracy of movie reviews using the integrated CNN-LSTM model.
Purpose : Cyber-Knife tumor tracking system, based on the correlation relationship between the position of a tumor which moves in response to the real time respiratory cycle signal and respiration was obtained by the LED marker attached to the outside of the patient, the location of the tumor to predict in advance, the movement of the tumor in synchronization with the therapeutic device to track real-time tumor, is a system for treating. The purpose of this study, in the cyber knife tumor tracking radiation therapy, trying to evaluate the accuracy of tumor tracking radiation therapy system due to the change in the form of unpredictable sudden breathing due to cough and sleep. Materials and Methods : Breathing Log files that were used in the study, based on the Respiratory gating radiotherapy and Cyber-knife tracking radiosurgery breathing Log files of patients who received herein, measured using the Log files in the form of a Sinusoidal pattern and Sudden change pattern. it has been reconstituted as possible. Enter the reconstructed respiratory Log file cyber knife dynamic chest Phantom, so that it is possible to implement a motion due to respiration, add manufacturing the driving apparatus of the existing dynamic chest Phantom, Phantom the form of respiration we have developed a program that can be applied to. Movement of the phantom inside the target (Ball cube target) was driven by the displacement of three sizes of according to the size of the respiratory vertical (Superior-Inferior) direction to the 5 mm, 10 mm, 20 mm. Insert crosses two EBT3 films in phantom inside the target in response to changes in the target movement, the End-to-End (E2E) test provided in Cyber-Knife manufacturer depending on the form of the breathing five times each. It was determined by carrying. Accuracy of tumor tracking system is indicated by the target error by analyzing the inserted film, additional E2E test is analyzed by measuring the correlation error while being advanced. Results : If the target error is a sine curve breathing form, the size of the target of the movement is in response to the 5 mm, 10 mm, 20 mm, respectively, of the average
The present study is an attempt to solve the basic problems involved in the control of the Sclerotium disease. The biologic stranis of Sclerotium rolfsii Sacc., pathogen of Sclerotium disease of Magnolia kobus, were differentiated, and the effects of vitamins, various nitrogen and carbon sources on its mycelial growth and sclerotial production have been investigated. In addition the relationship between the cultural filtrate of Penicillium sp. and the growth of Sclerotium rolfsii, the tolerance of its mycelia or sclerotia to moist heat or drought and to Benlate (methyl-(butylcarbamoy 1)-2-benzimidazole carbamate), Tachigaren (3-hydroxy-5-methylisoxazole) and other chemicals were also clarified. The results are summarizee as follows: 1. There were two biologic strains, Type-l and Type-2 among isolates. They differed from each other in the mode of growth and colonial appearance on the media, aversion phenomenon and in their pathogenicity. These two types had similar pathogenicity to the Magnolia kobus and Robinia pseudoacasia, but behaved somewhat differently to the soybaen and cucumber, the Type-l being more virulent. 2. Except potassium nitrite, sodium nitrite and glycine, all of the 12 nitrogen sources tested were utilized for the mycelial growth and sclerotial production of this fungus when 10r/l of thiamine hydrochloride was added in the culture solution. Considering the forms of nitrogen, ammonium nitrogen was more available than nitrate nitrogen for the growth of mycelia, but nitrate nitrogen was better for sclerotia formation. Organic nitrogen showed different availabilities according to compounds used. While nitrite nitrogen was unavailable for both mycelial growth and sclerotial formation whether thiamine hydrochlioride was added or not. 3. Seven kinds of carbon sources examined were not effective in general, as long as thiamine hydrochloride was not added. When thiamine hydrochloride was added, glucose and saccharose exhibited mycelial growth, while rnaltose and soluble starch gave lesser, and xylose, lactose, and glycine showed no effect at all,. In the sclerotial production, all the tested carbon sources, except lactose, were effective, and glucose, maltose, saccharose, and soluble starch gave better results. 4. At the same level of nitrogen, the amount of mycelial growth increased as more carbon Sources were applied but decreased with the increase of nitrogen above 0.5g/1. The amount of sclerotial production decreased wi th the increase of carbon sources. 5. Sclerotium rolfsii was thiamine-defficient and required thiamine 20r/l for maximun growth of mycelia. At a higher concentration of more than 20r/l, however, mycelial growth decreased as the concentration increased, and was inhibited at l50r/l to such a degree of thiamine-free. 6. The effect of the nitrogen sources on the mycelial growth under the presence of thiamine were recognized in the decreasing order of