Shoffan Saifullah, Yuli Fauziah, Agus Sasmito Aribowo
(参考訳) 新型コロナウイルスのパンデミックの影響をすべてのグループが感じた。
この状況は不安を引き起こすが、これは誰にとっても悪い。
政府の役割は、これらの問題をその事業プログラムで解くことに非常に影響を与える。
また、公衆の不安を引き起こす多くの長所や短所もある。
そのため、公共の期待を高めることができる政府プログラムを改善するために不安を検出する必要がある。
本研究は、このパンデミックに対処する政府のプログラムに関するソーシャルメディアコメントに基づく不安検出に機械学習を適用した。
この概念は、netizensのポジティブなコメントとネガティブなコメントに基づく不安を検出するために、感情分析を採用する。
実装された機械学習方法は、k-nn, bernoulli, decision tree classifier, support vector classifier, random forest, xg-boostである。
使用したデータはYouTubeコメントをクロールした結果である。
使用されたデータは3211と1651の否定的データと肯定的データからなる4862のコメントだった。
負のデータは不安を識別し、正のデータは希望(不安ではない)を識別する。
機械学習は、カウントベクタライゼーションとTF-IDFの特徴抽出に基づいて処理される。
その結果、感情データは3889と973であり、最も精度の高いトレーニングは、ベクトル化数の特徴抽出とTF-IDFが84.99%、TF-IDFが82.63%であるランダム森林であった。
最良の精度テストはK-NN、最良のリコールはXG-Boostである。
したがって、ランダムフォレストは、ソーシャルメディアから誰かの不安に基づくデータを検出するのに最適である。
All groups of people felt the impact of the COVID-19 pandemic. This situation
triggers anxiety, which is bad for everyone. The government's role is very
influential in solving these problems with its work program. It also has many
pros and cons that cause public anxiety. For that, it is necessary to detect
anxiety to improve government programs that can increase public expectations.
This study applies machine learning to detecting anxiety based on social media
comments regarding government programs to deal with this pandemic. This concept
will adopt a sentiment analysis in detecting anxiety based on positive and
negative comments from netizens. The machine learning methods implemented
include K-NN, Bernoulli, Decision Tree Classifier, Support Vector Classifier,
Random Forest, and XG-boost. The data sample used is the result of crawling
YouTube comments. The data used amounted to 4862 comments consisting of
negative and positive data with 3211 and 1651. Negative data identify anxiety,
while positive data identifies hope (not anxious). Machine learning is
processed based on feature extraction of count-vectorization and TF-IDF. The
results showed that the sentiment data amounted to 3889 and 973 in testing, and
training with the greatest accuracy was the random forest with feature
extraction of vectorization count and TF-IDF of 84.99% and 82.63%,
respectively. The best precision test is K-NN, while the best recall is
XG-Boost. Thus, Random Forest is the best accurate to detect someone's anxiety
based-on data from social media.
The machine learning methods implemented include K-NN, Bernoulli, Decision Tree Classifier, Support Vector Classifier, Random Forest, and XG-boost.
実装された機械学習方法は、k-nn, bernoulli, decision tree classifier, support vector classifier, random forest, xg-boostである。
0.71
The data sample used is the result of crawling YouTube comments.
使用したデータはYouTubeコメントをクロールした結果である。
0.70
The data used amounted to 4862 comments consisting of negative and positive data with 3211 and 1651.
使用されたデータは3211と1651の否定的データと肯定的データからなる4862のコメントだった。
0.64
Negative data identify anxiety, while positive data identifies hope (not anxious).
負のデータは不安を識別し、正のデータは希望(不安ではない)を識別する。
0.57
Machine learning is processed based on feature extraction of count-vectorization and TFIDF.
機械学習は、カウントベクタライゼーションとTFIDFの特徴抽出に基づいて処理される。
0.71
The results showed that the sentiment data amounted to 3889 and 973 in testing, and training with the greatest accuracy was the random forest with feature extraction of vectorization count and TF-IDF of 84.99% and 82.63%, respectively.
These government programs have pros and cons, which can be seen in various news and social media.
これらの政府のプログラムには賛否両論があり、様々なニュースやソーシャルメディアで見ることができる。
0.71
Comments and feedback on social media are highly visible and indicate whether there is anxiety or not.
ソーシャルメディアに対するコメントやフィードバックは目立っており、不安があるかどうかを示している。
0.72
Everyone can make comments and feedback from all walks of life.
誰もが人生のあらゆる段階からコメントやフィードバックをすることができる。
0.72
News in the Covid-19 pandemic, such as details of the spread, death rates, and government programs for handling it, has a massive impact on the community's psyche, such as anxiety and panic [30].
Anxiety can be detected using the concept of sentiment analysis [2, 27, 46], especially in text processing.
特にテキスト処理において感情分析[2, 27, 46]の概念を用いて不安を検出することができる。
0.84
The study explains that sentiment analysis is used to detect anxiety based on one's social media, both from sharing information and feedback (comments) [25].
However, along with the development of technology and the concept of text mining, anxiety detection can be done quickly and precisely based on training data using machine learning.
This concept process used an analysis of a person's sentiment in responding to the pandemic and government programs in dealing with the COVID19 pandemic.
In its identification, this study analyzes emotions in the form of detection of fanaticism in the document text.
本研究は,文書中の狂信性を検出する形で感情を解析する。
0.59
In the detection of fanaticism, the concept of fanaticism is divided into three, namely non-fanaticism, Code Attitude fanaticism, and Code Red Fanaticism.
The implementation of machine learning methods in the classification process reaches 96%.
分類プロセスにおける機械学習手法の実装は96%に達する。
0.86
The methods used are C4.5, RIPPER, and PART.
使用される方法はC4.5、RIPPER、Partである。
0.64
These results are included in the structured domain [4].
これらの結果は構造化ドメインに含まれます [4]。
0.86
After that, this research was developed using the feature extraction method.
その後,特徴抽出法を用いて本研究を行った。
0.78
The feature extraction methods used include TF-IDF, Support Vector Machine, and Naïve Bayesian.
特徴抽出にはtf-idf、サポートベクターマシン、ナイーブ・ベイジアンが含まれる。
0.61
Implementation with the feature extraction method yields an accuracy of 82.1% [5].
特徴抽出法による実装は82.1%[5]の精度が得られる。
0.88
Research in sentiment analysis can detect hate speech based on data from Facebook.
感情分析の研究は、facebookのデータに基づいてヘイトスピーチを検出することができる。
0.57
This research belongs to unstructured sentiment analysis [40].
この研究は非構造化感情分析に属する[40].
0.89
The application of the Support Vector Machines (SVM) method and a particular Recurrent Neural Network, namely Long Short Term Memory (LSTM), were tested for performance in emotion recognition.
Support Vector Machines (SVM) 法と,特定のリカレントニューラルネットワーク,すなわちLong Short Term Memory (LSTM) を適用し,感情認識の性能を検証した。
0.75
Emotions identified by this method are positive and negative emotions.
この方法で同定された感情はポジティブ感情とネガティブ感情である。
0.54
The study resulted in the effectiveness of both approaches in the Italian language domain.
この研究は、イタリア語領域における両方のアプローチの有効性をもたらした。
0.71
Other studies also apply the lexicon method and a combination of sentiment dictionaries and sentiment corpus in the sentiment detection process.
This detection is based on the negative sentiment, which is defined as displeasing utterances related to religion, gender, and specific ethnicity [43].
Besides, related research detects sentiment using the two-step method [11].
さらに、関連研究は2段階法による感情を検出する[11]。
0.76
Likewise, in other studies, sentiment analysis in the detection process is based on individual objects oriented towards race, nationality, and religion [15].
Analysis of sentiment detection is automatically processed using natural language processing [32] using text data on social media, including analytical sentiment detection on Facebook [40].
The detection of one's expectations is also carried out using social media data in the political domain [8].
また、政治領域におけるソーシャルメディアデータを用いて、期待の検出を行う [8]。
0.64
The data used is data from Twitter.
使用されるデータはTwitterのデータだ。
0.84
Researches with the theme of detection of mass anxiety and fear have also been carried out in various circumstances, namely during political wars [10, natural earthquake disasters [41], and in Youtube video comments [9].
There are four main steps taken in this research: data collection, system analysis and concepts, and system design.
この研究には、データ収集、システム分析と概念、システム設計の4つの主要なステップがある。
0.83
Data collection is done by looking for literature studies to get an overview and literacy related to machine learning and text mining following the problems being solved.
The development of this research is the application of feature extraction methods with Count-Vectorization and TF-IDF and machine learning methods using K-NN, Bernoulli, Decision Tree Classifier, Support Vector Classifier, Random Forest, and XG-Boost.
The prototyping process uses three main stages in text processing: preprocessing, emotion detection based on sentiment analysis, and cross-validation testing.
Meanwhile, TF-IDF is a method used to calculate the weight of each word that is commonly used.
一方、TF-IDFは一般的に使用される単語の重みを計算する方法である。
0.73
The method has advantages in several factors, such as efficiency, ease, and accuracy.
この方法は、効率、容易性、正確性などのいくつかの要因において利点がある。
0.56
This method aims to calculate the value of the Term Frequency (TF) and Inverse Document Frequency (IDF) of each token (word) of each document in the corpus.
Alternatively, the TF-IDF method serves to find out how often the word appears.
あるいは、TF-IDF法は単語の出現頻度を調べるのに役立つ。
0.73
It is the first step in the machine learning process stages, according to Figure 2.
図2に示すように、機械学習プロセスのステージにおける最初のステップである。
0.85
Machine learning will process data based on the results of feature extraction.
機械学習は特徴抽出の結果に基づいてデータを処理する。
0.89
This machine learning process is a concept used to detect sentiments and emotions based on the text whose comments are processed.
この機械学習プロセスは、コメントを処理するテキストに基づいて感情や感情を検出するための概念である。
0.83
As stated in the development of previous studies, this study uses five machine learning methods for the detection process.
先行研究の展開で述べられているように,本研究は検出プロセスに5つの機械学習手法を用いる。
0.81
The result will be a training and testing process.
その結果はトレーニングとテストのプロセスになります。
0.81
The results of the training and test data are calculated for its accuracy using cross-validation.
クロスバリデーションを用いて、トレーニングデータとテストデータの結果を精度良く算出する。
0.71
Apart from accuracy, other calculations use precision and recall functions.
精度は別として、他の計算では精度とリコール関数を用いる。
0.61
YoutubeVideoComments PreprocessingTokeniz ingFilteringStemming TaggingEmoticonConve rtto textComment DataCrawlingClean Data
YoutubeVideoComments PreprocessingTokeniz ingFilteringStemming TaggingEmoticonConve rtto textComment DataCrawlingClean Data
0.09
英語(論文から抽出)
日本語訳
スコア
4 Figure 2. The stages of Machine Learning and the results
4 図2。 機械学習の段階と結果
0.73
Machine learning uses the K-NN (K-Nearest Neighbor) method, Bernoulli, Decision Tree Classifier, Support Vector Classifier, Random Forest, and XG-Boost.
機械学習はK-NN(K-Nearest Neighbor)メソッド、ベルヌーイ、決定木分類器、サポートベクトル分類器、ランダムフォレスト、XG-Boostを使用する。 訳抜け防止モード: 機械学習はK-NN(K-Nearest Neighbor )メソッドを使用する。 Bernoulli, Decision Tree Classifier, Support Vector Classifier, Random Forest そして XG - Boost 。
0.87
KNN is used in the classification process by finding the closest K match based on training data [28].
Furthermore, the closest match label is used in the prediction.
さらに、最も近いマッチングラベルが予測に使用される。
0.69
In general, the distance used is the euclidean distance in finding the closest match.
一般に、最も近い一致を見つける際に使用される距離はユークリッド距離である。
0.79
The second method is Bernoulli.
2つ目の方法はベルヌーイである。
0.41
The model is used only to ignore the number of occurrences [21, 42].
このモデルは[21, 42]の発生回数を無視するためにのみ使用される。
0.81
Bernoulli's Naïve Bayes classification is a model that specifies that a document is represented by a binary attribute vector that indicates words that appear and do not appear in the document [22].
The decision tree is a method of classification and pattern prediction from data.
決定木は、データからの分類とパターン予測の方法である。
0.83
It describes the relationship of the attribute variable x and target y in the form of a tree.
属性変数 x とターゲット y の関係を木の形で記述する。
0.54
The decision tree has a structure like a flowchart, where the internal node is a test of the attribute variable (not the leaf / outermost), each branch is the test result, and the outer node is the leaf, which is the label [37].
The concept of classification with the Support Vector Machine is to find the best hyperplane that functions as a separator of two data classes.
Support Vector Machineによる分類の概念は、2つのデータクラスのセパレータとして機能する最高の超平面を見つけることである。
0.82
The Support Vector Machine can work on high-dimensional datasets using kernel tricks.
Support Vector Machineはカーネルのトリックを使って高次元のデータセットを扱うことができる。
0.64
The Support Vector Machine only uses a few selected data points that contribute (support vector) to form a model used in the classification process [31].
Random Forest (RF) is a classification algorithm that can be used to identify sentiment analysis [36, 20] and emotion analysis [16, 18] in numbers significant data.
This algorithm mimics the behavior of Random forest in its tree creation and is also combined with gradient descent / boosting.
このアルゴリズムは、木の形成におけるランダム森林の挙動を模倣し、勾配降下/隆起と組み合わせる。
0.75
Gradient Boosting is a machine learning concept in solving regression problems whose classification produces a predictive model in a weak ensemble prediction model [47].
Clean Opinion DataCross ValidationResultMach ine LearningFeature Extraction
clean opinion datacross validationresultmach ine learning機能抽出
0.83
英語(論文から抽出)
日本語訳
スコア
5 measured by (3).
5 (3)で測定した。
0.77
Precision is the number of opinion samples "true" label, which means positive sentiment and is divided by the total number of positive sentiment classification samples.
FINDINGS AND DISCUSSION In this study, there are results and discussions related to the experiments that have been carried out.
発見と議論 本研究では,これまで実施されてきた実験に関する結果と議論について述べる。
0.61
The details obtained in this experiment include explaining the dataset, the results, and discussions related to the anxiety detection process based on sentiment analysis of social media data.
Setuju nih.. kebanyakan ngibul nasih gratisan aja..
ケバニャカン・ナシフ・グラティザン・アジャ(Setuju nih. kebanyakan ngibul nasih gratisan aja.)
0.52
Ya gk pp di syukuri aj ya
gk pp di syukuri aj ya
0.70
Alhamdulillah saya bisa
Alhamdulillah saya bisa
0.85
Sabar nggeh, insyaallah rezekinya ada terus.
sabar nggeh, insyaallah rezekinya ada terus。
0.72
Aamiin Almhmdllh sdh dpat akses dan sdh dpat nmer tokennya...
阿弥院 almhmdllh sdh dpat akses dan sdh dpat nmer tokennya...
0.68
Category Negative Negative Negative
カテゴリー 負 負 負
0.66
Negative Negative Positive Positive Positive
負 負 陽性 陽性 陽性
0.59
Positive
陽性
0.55
英語(論文から抽出)
日本語訳
スコア
6 10 Coba terus sampe keluar tokennya gan....saya semalam gitu...alhamdulillah sudah dapat
6 10 Coba terus Sampe keluar tokennya gan....saya semalam gitu...alhamdulillah sudah dapat
0.86
Positive B. Experiment Results and Discussion
陽性 B。 実験結果と議論
0.69
The experimental process was carried out using crawled data.
実験はクロールデータを用いて実施した。
0.67
The results of crawling data from YouTube comments are presented in Table 2 as a sample.
YouTubeコメントからクロールしたデータは、テーブル2にサンプルとして表示される。
0.71
The data is pre-processed to obtain clean data for further processing.
データは前処理され、さらに処理するためのクリーンデータを取得する。
0.64
This process fully uses the Python programming language.
このプロセスはPythonプログラミング言語を完全に使用する。
0.78
Classification result data was carried out using two data groupings, namely negative and positive.
分類結果データは2つのデータグループ(負と正)を用いて行われた。
0.77
Python implementation uses a design as in the research in Figure 1 and Figure 2 for the process flow.
pythonの実装では、図1や図2にあるように、プロセスフローにデザインを使用します。
0.71
The process begins by reading the data and then preprocessing.
プロセスはデータの読み込みと前処理から始まります。
0.81
Preprocessing aims to clean data.
プリプロセッシングはデータのクリーン化を目標とする。
0.51
The resulting data is a core that has no noise.
その結果得られたデータは、ノイズのないコアになります。
0.67
This process is carried out with several processes, including: tokenizing, filtering (slank word conversion, remove the number, remove stopword, remove the figure, remove duplicate), stemming, emoticon conversion.
In the conversion process, emoticons have rules to improve data cleanliness.
変換プロセスでは、エモティコンはデータのクリーン化を改善するためのルールを持つ。
0.52
Additional rules besides the process are as follows:
プロセス以外のルールは以下の通りである。
0.75
a.There is a single figure word; it is necessary to add the word "support".
a. 一つの図形語が存在し、" Support" という単語を追加する必要がある。
0.72
The result is positive sentiment
結果はポジティブな感情です
0.80
and emotional trust = 1 b.There is an exclamation mark, then the emotion of trust = 2 c.In one dominant capital letter sentence, angry emotions = 1 d.There is an angry emotion exclamation mark = 2 Here is a coding sample in Python to display the conversion of emotions from comment text.
感情的信頼 = 1 b) 宣言マークがあるならば,信頼の感情 = 2 c. 支配的な大文字文では,怒りの感情 = 1 d. 怒りの感情の宣言マーク = 2 コメントテキストからの感情の変換を表示するためのPythonのコーディングサンプルがある。
0.78
The conversion result is a complete sentence without emoticons.
変換結果はエモティコンのない完全文である。
0.63
This is because the emoticon is converted into a complete sentence, what changes is the emoticon becomes the sentence ".... face_with_tears_of_j oy.
(b) (a) calculations based on the Random Forest method
(b) (a) ランダムフォレスト法による計算
0.76
Based on Figure 5, the use of feature extraction methods and machine learning in anxiety detection is implemented in Figure 6.
図5に基づいて、不安検出における特徴抽出法と機械学習の使用を図6に実装する。
0.79
These implementation results are examples of applying the feature extraction method using Count-Vectorization, and the machine learning method using the random forest method.
The validation process uses the application of formulas (1), (2), and (3), which are coded in Python with the functions provided in the python library.
Based on the matrix results in Figure 6, the results of the test results for each feature extraction method and machine learning are shown in Table 3.
図6のマトリックス結果に基づいて、各特徴抽出方法と機械学習のためのテスト結果の結果を表3に示す。
0.77
The highest accuracy result is the random forest method based on each feature extraction Count-Vectorization and TF-IDF with successive values are 84.99% and 82.63%.
Apart from Random Forest, the application of machine learning methods with an accuracy of more than 80% and are at least balanced in calculating precision and recall can be considered.
The three methods apart from having an accuracy of more than 80%, the precision and recall can be balanced.
3つの方法は精度が80%以上であるのとは別に、精度とリコールのバランスをとることができる。
0.75
The precision and recall values were more than 60.
精度とリコール値は60以上であった。
0.72
Each result had differences that were not that far off, with a maximum difference of about 20% of the results.
それぞれの結果にはそれほど遠くない違いがあり、最大で20%の差があった。
0.68
英語(論文から抽出)
日本語訳
スコア
8 Figure 7. The graph of the Machine Learning method is based on cross-validation calculations
8 図7。 クロスバリデーション計算に基づく機械学習手法のグラフ
0.73
Tabel 3. Results of the calculation of cross-validation from the Machine Learning Methods (such as K-NN, Bernoulli, Decision Tree, Support Vector Classifier, Random Forest and XG-Boost)
タベル3。 K-NN, Bernoulli, Decision Tree, Support Vector Classifier, Random Forest, XG-Boost などの機械学習手法によるクロスバリデーションの計算結果
0.65
Machine Learning Count-Vectorization
機械学習 Count-Vectorization
0.68
Accuracy Precission K-NN Bernoulli Decission Tree Support Vector Classifier Random Forest XG-Boost
精度 精度 K-NN Bernoulli Deission Tree Support Vector Classifier Random Forest XG-Boost
Cross-Validation Testing, especially the level of accuracy, the random forest has an accuracy rate of close to 85% with Count-Vectorization feature extraction compared to all the machine learning methods used.
Even so, the accuracy of applying machine learning methods with an accuracy of more than 80% includes Bernoulli, Decision Tree, Support Vector Classifier, and Random Forest.
それでも、80%以上の精度で機械学習手法を適用する精度には、bernolli、 decision tree、 support vector classificationifier、random forestが含まれる。
0.70
Thus, Random Forest can be the best reference for applying machine learning methods to detect anxiety based on social media data.
REFERENCES Ahmad, A. R. and Murad, H. R. (2020) ‘The Impact of Social Media on Panic During the COVID-19 Pandemic in Iraqi Kurdistan: Online Questionnaire Study’, Journal of Medical Internet Research, 22(5), p. e19556.
Ahmad, A. R. and Murad, H. R. (2020) 'The Impact of Social Media on Panic during the COVID-19 Pandemic in Iraqi Kurdistan: Online Questionnaire Study', Journal of Medical Internet Research, 22(5), p. e 19556。
0.90
doi: 10.2196/19556.
10.2196/19556。
0.66
Aladağ, A. E. et al (2018) ‘Detecting Suicidal Ideation on Forums: Proof-of-Concept Study’, Journal of
Alada', A. E. et al (2018) 'Detecting Suicidal Ideation on Forums: Proof-of-Concept Study', Journal of Journal
0.86
Medical Internet Research, 20(6), p. e215.
Medical Internet Research, 20(6), p. e215。
0.88
doi: 10.2196/jmir.9840.
10.2196/jmir.9840
0.54
Almonayyes, A.
Almonayyes, A。
0.80
(2006) ‘Multiple Explanations Driven Naive Bayes Classifier.’, Journal off Universal
(2006) 'Multiple Explanations Driven Naive Bayes Classifier', Journal off Universal
0.78
Computer Science, 12(2), pp.
コンピュータ科学、12(2), pp。
0.66
127–139. Almonayyes, A.
127–139. Almonayyes, A。
0.75
(2016) ‘Classifying Documents By Integrating Contextual Knowledge With Boosting’, in
(2016)「文脈知識とブースティングの統合による文書の分類」
0.65
International Conference on Artificial Intelligence and Computer Science, pp.
人工知能とコンピュータサイエンスに関する国際会議, pp.
0.75
28–29. Almonayyes, A.
28–29. Almonayyes, A。
0.75
(2017) ‘Tweets Classification Using Contextual Knowledge And Boosting’, International
(2017)'Tweets Classification using Contextual Knowledge and Boosting', International
0.72
Journal of Advances in Electronics and Computer Science, (4), pp.
Journal of Advances in Electronics and Computer Science, (4), pp.
0.80
87–92. Asyary, A. and Veruswati, M. (2020) ‘Sunlight exposure increased Covid-19 recovery rates: A study in the central pandemic area of Indonesia’, Science of The Total Environment, 729, p. 139016. doi: 10.1016/j.scitotenv. 2020.139016.
87–92. Asyary, A. and Veruswati, M. (2020) ‘Sunlight exposure increased Covid-19 recovery rate: A study in the central pandemic area of Indonesia’, Science of The Total Environment, 729, pp. 139016. doi: 10.1016/j.scitotenv. 2020.139016.
0.78
Bhati, R. (2020) ‘Sentiment analysis a deep survey on methods and approaches’.
bhati, r. (2020) ‘sentiment analysis a deep survey on methods and approach’。
0.81
Science & Engineering Research Support society.
科学と工学 研究支援団体。
0.71
英語(論文から抽出)
日本語訳
スコア
9 Calderón-Monge, E. (2017) ‘Twitter to Manage Emotions in Political Marketing’, Journal of Promotion
9 Calderón-Monge, E. (2017) 'Twitter to Managage Emotions in Political Marketing', Journal of promotion
0.87
Management. Taylor & Francis, 23(3), pp.
管理。 Taylor & Francis, 23(3), pp。
0.74
359–371. doi: 10.1080/10496491.201 7.1294870.
359–371. 1080/10496491.2017.1 294870
0.60
Chen, Y. L., Chang, C. L. and Yeh, C. S. (2017) ‘Emotion Classification of YouTube Videos’, Decision
Chen, Y. L., Chang, C. L. and Yeh, C. S. (2017) ‘Emotion Classification of YouTube Videos’, Decision
0.99
Support Systems, 101, pp.
サポートシステム, 101, pp。
0.78
40–50. doi: 10.1016/j.dss.2017.0 5.014.
40–50. doi: 10.1016/j.dss.2017.0 5.014。
0.52
Chin, D., Zappone, A. and Zhao, J.
Chin, D., Zappone, A. and Zhao, J。
0.87
(2016) ‘Analyzing Twitter Sentiment of the 2016 Presidential
2016年)「2016年大統領選挙のTwitter感の分析」
0.76
Candidates’, Applied Informatics and Technology Innovation Conference (AITIC 2016).
Candidates’, Applied Informatics and Technology Innovation Conference (AITIC 2016)。
0.78
Djuric, N. et al (2015) ‘Hate Speech Detection with Comment Embeddings’, pp.
Djuric, N. et al (2015) ‘Hate Speech Detection with Comment Embeddings’, pp。
0.86
29–30. Fauziah, Y., Saifullah, S. and Aribowo, A. S. (2020) ‘Design Text Mining for Anxiety Detection using Machine Learning based-on Social Media Data during COVID-19 pandemic’, in Proceeding of LPPM UPN “Veteran” Yogyakarta Conference Series 2020–Engineering and Science Series, pp.
29–30. Fauziah, Y., Saifullah, S. and Aribowo, A.S. (2020) ‘Design Text Mining for Anxiety Detection using Machine Learning based-on Social Media Data during COVID-19 pandemic’ in Proceeding of LPPM UPN “Veteran” Yogyakarta Conference Series 2020–Engineering and Science Series, pp. ^ 原文(投稿日:2012/01/28) 訳抜け防止モード: 29–30. Fauziah, Y., Saifullah, S. and Aribowo, A.S. (2020 ) 'Design Text Mining for Anxiety Detection using Machine Learning based on Social Media data during COVID-19 pandemic ' In Proceeding of LPPM UPN “ Veteran ” Yogyakarta Conference Series 2020–Engineering and Science Series,pp.
0.82
253–261. doi: 10.31098/ess.v1i1.11 7.
253–261. doi: 10.31098/ess.v1i1.11 7。
0.53
Georganos, S. et al.
Georganos, S. et al。
0.94
(2018) ‘Very High Resolution Object-Based Land Use–Land Cover Urban Classification Using Extreme Gradient Boosting’, IEEE Geoscience and Remote Sensing Letters, 15(4), pp.
Gitari, N. D. et al (2015) ‘A Lexicon-based Approach for Hate Speech Detection’, International Journal of
Gitari, N. D. et al (2015) 'A Lexicon-based Approach for Hate Speech Detection', International Journal.
0.91
Multimedia and Ubiquitous Engineering, 10(4), pp.
マルチメディアとユビキタスエンジニアリング, 10(4), pp。
0.67
215–230. doi: 10.14257/ijmue.2015. 10.4.21.
215–230. 10.14257/ijmue.2015. 10.4.21
0.52
Gokulakrishnan, B. et al.
Gokulakrishnan, B. et al。
0.93
(2012) ‘Opinion mining and sentiment analysis on a Twitter data stream’, in International Conference on Advances in ICT for Emerging Regions (ICTer2012).
(2012)「Twitterデータストリームのオピニオンマイニングと感情分析」は、ICT for Emerging Regions(ICTer2012)の国際会議において発表された。
0.81
IEEE, pp. 182–188.
IEEE, pp。 182–188.
0.76
doi: 10.1109/ICTer.2012.6 423033.
10.1109/ICTer.2012.6 423033
0.44
Hamzah, F. A. B. et al.
ハンザ、F.A。 b. et al. など。
0.61
(2020) ‘CoronaTracker: worldwide COVID-19 outbreak data analysis and
(2020年)「コロナトラッカー」の世界的流行データ分析
0.65
prediction’, Bull World Health Organ, 1, p. 32.
とBull World Health Organ, 1, pp. 32。
0.55
Jayalekshmi, J. and Mathew, T. (2017) ‘Facial expression recognition and emotion classification system for sentiment analysis’, in 2017 International Conference on Networks & Advances in Computational Technologies (NetACT).
Jayalekshmi, J. and Mathew, T. (2017) 'Facial expression recognition and emotion classification system for sentiment analysis' in 2017 International Conference on Networks & Advances in Computational Technologies (NetACT)
0.83
IEEE, pp. 1–8.
IEEE, pp。 1–8.
0.76
doi: 10.1109/NETACT.2017. 8076732.
10.1109/NETACT.2017. 8076732
0.43
Kléma, J. and Almonayyes, A.
Kléma, J. and Almonayyes, A.
0.94
(2006) ‘Automatic Categorization of Fanatic Text Using random Forests’,
(2006年)「ランダムフォレストを用いたファナティックテキストの自動分類」
0.68
Kuwait Journal of Science and Engineering, 33(2), pp.
Kuwait Journal of Science and Engineering, 33(2), pp。
0.74
1–18. Kumar, S., Yadava, M. and Roy, P. P. (2019) ‘Fusion of EEG response and sentiment analysis of products Information Fusion, 52, pp.
1–18. Kumar, S., Yadava, M. and Roy, P. P. (2019) 'Fusion of EEG response and sentiment analysis of products Information Fusion, 52, pp。
Manning, C., Raghavan, P. and Schütze, H. (2009) An Introduction to Information Retrieval.
Manning, C., Raghavan, P. and Schütze, H. (2009) An Introduction to Information Retrieval
0.86
Cambridge: Cambridge University Press.
ケンブリッジ ケンブリッジ大学出版局。
0.52
McCallum, A. and Nigam, K. (1998) ‘A comparison of event models for naive bayes text classification’,
mccallum, a. and nigam, k. (1998) 'a comparison of event models for naive bayes text classification' (英語)
0.87
in AAAI-98 workshop on learning for text categorization, pp.
In AAAI-98 Workshop on Learning for text categorization, pp。
0.90
41–48. Nanur, F. N., Halu, S. A. N. and Juita, E. (2020) ‘Evaluasi ketersediaan fasilitas kesehatan yang memadai
41–48. Nanur, F. N., Halu, S. A. N. and Juita, E. (2020) ‘Evaluasi ketersediaan fasilitas kesehatan yang memadai
0.84
terhadap pencapaian revolusi kia di Manggarai’, Jurnal Kebidanan, 12(01), pp.
terhadap pencapaian revolusi kia di manggarai', jurnal kebidanan, 12(01), pp。
0.72
80–92. Ni, M. Y. et al (2020) ‘Mental Health, Risk Factors, and Social Media Use During the COVID-19 Epidemic and Cordon Sanitaire Among the Community and Health Professionals in Wuhan, China: CrossSectional Survey’, JMIR Mental Health, 7(5), p. e19009.
80–92. Ni, M. Y. et al (2020) 'Mental Health, Risk Factors and Social Media Use during the COVID-19 Epidemic and Cordon Sanitaire among the Community and Health Professionals in the Wuhan in China: Cross Sectional Survey, JMIR Mental Health, 7(5), p. e19009。
0.80
doi: 10.2196/19009.
10.2196/19009。
0.66
Osadchiy, V., Mills, J. N. and Eleswarapu, S. V. (2020) ‘Understanding Patient Anxieties in the Social Media Era: Qualitative Analysis and Natural Language Processing of an Online Male Infertility Community’, Journal of Medical Internet Research, 22(3), p. e16728.
Osadchiy, V., Mills, J. N. and Elesharapu, S. V. (2020) ‘Understanding patient Anxieties in the Social Media Era: Qualitative Analysis and Natural Language Processing of an Online Male Infertility Community’, Journal of Medical Internet Research, 22(3), p. e16728。
0.91
doi: 10.2196/16728.
10.2196/16728
0.71
Rabbimov, I. et al (2020) ‘Investigating the Effect of Emoji in Opinion Classification of Uzbek Movie
Rabbimov, I. et al (2020)'investigating the Effect of Emoji in Opinion Classification of Uzbek Movie
Ragini, J. R., Anand, P. M. R. and Bhaskar, V. (2018) ‘Big data analytics for disaster response and recovery through sentiment analysis’, International Journal of Information Management, 42, pp.
Ragini, J. R., Anand, P. M. R. and Bhaskar, V. (2018) ‘Big Data Analytics for disaster response and recovery through sentiment analysis’, International Journal of Information Management, 42, pp。
0.93
13–24. doi: 10.1016/j.ijinfomgt. 2018.05.004.
13–24. doi: 10.1016/j.ijinfomgt. 2018.05.004。
0.52
Rezwanul, M., Ali, A. and Rahman, A.
Rezwanul, M., Ali, A. and Rahman, A.
0.92
(2017) ‘Sentiment Analysis on Twitter Data using KNN and SVM’, and Applications, 8(6).
(2017) 'Sentiment Analysis on Twitter Data using KNN and SVM', and Applications, 8(6)。
0.77
doi: Journal of Advanced Computer Science
Doi: journal of advanced computer science(英語)
0.60
International 10.14569/IJACSA.2017 .080603.
国際10.14569/IJACSA.2017 .080603
0.40
Saifullah, S. (2019) ‘Fuzzy-AHP approach using Normalized Decision Matrix on Tourism Trend Ranking
Saifullah, S. (2019)'Fuzzy-AHP approach using normalized Decision Matrix on Tourism Trend Ranking
0.93
based-on Social Media’, Jurnal Informatika, 13(2), pp.
based-on social media', jurnal informatika, 13(2), pp。
0.78
16–23. doi: 10.26555/jifo.v13i2. a15268.
16–23. doi: 10.26555/jifo.v13i2. a15268。
0.52
Saputra, T. A.
Saputra, T. A。
0.93
(2020) ‘Bentuk Kecemasan dan Resiliensi Mahasiswa Pascasarjana Aceh-Yogyakarta Dalam
(2020年)「Bentuk Kecemasan dan Resiliensi Mahasiswa Pascasarjana Aceh-Yogyakarta Dalam」
0.87
Menghadapi Pandemi Covid-19’, Jurnal Bimbingan Dan Konseling Ar-Rahman, 6(1), pp.
Menghadapi Pandemi Covid-19', Jurnal Bimbingan Dan Konseling Ar-Rahman, 6(1), pp。
0.90
55–61.
55–61.
0.71
英語(論文から抽出)
日本語訳
スコア
10 Sari, B. W. and Haranto, F. F. (2019) ‘Implementasi Support Vector Machine Untuk Analisis Sentimen Pengguna Twitter Terhadap Pelayanan Telkom dan Biznet’, Jurnal Pilar Nusa Mandiri, 15(2), pp.
10 Sari, B. W. and Haranto, F. F. (2019) ‘Implementasi Support Vector Machine Untuk Analisis Sentimen Pengguna Twitter Terhadap Pelayanan Telkom dan Biznet’, Jurnal Pilar Nusa Mandiri, 15(2), pp。
0.88
171– 176. doi: 10.33480/pilar.v15i2 .699.
171–176. doi: 10.33480/pilar.v15i2 .699
0.39
Schmidt, A. and Wiegand, M. (2017) ‘A Survey on Hate Speech Detection using Natural Language Processing’, in Fifth International Workshop on Natural Language Processing for Social Media, pp.
Schmidt, A. and Wiegand, M. (2017) 'A Survey on Hate Speech Detection using Natural Language Processing' in Fifth International Workshop on Natural Language Processing for Social Media, pp.
0.86
1– 10. doi: 10.18653/v1/W17-1101 .
10.18653/v1/W17-1101 。
0.48
Setiati, S. and Azwar, M. K. (2020) ‘COVID-19 and Indonesia’, Acta medica Indonesiana, 52(1), pp.
Setiati, S. and Azwar, M. K. (2020) ‘COVID-19 and Indonesia’, Acta medica Indonesiana, 52(1), pp。
0.95
84– 89. Available at: https://www.scopus.c om/record/display.ur i?eid=2-s2.0-85083414691&a mp;origin=inward.
Silalahi, D. E. and Ginting, R. R. (2020) ‘Strategi Kebijakan Fiskal Pemerintah Indonesia Untuk Mengatur Penerimaan dan Pengeluaran Negara Dalam Menghadapi Pandemi Covid-19’, Jesya (Jurnal Ekonomi & Ekonomi Syariah), 3(2), pp.
Silalahi, D. E. and Ginting, R. R. (2020) ‘Strategi Kebijakan Fiskal Pemerintah Indonesia Untuk Mengatur Penerimaan dan Pengeluaran Negara Dalam Menghadapi Pandemi Covid-19’, Jesya (Jurnal Ekonomi & Ekonomi Syariah), 3(2), pp。
0.92
156–167. Somawati, A. V. et al (2020) Bali vs COVID-19: Book Chapters.
156–167. Somawati, A. V. et al (2020) Bali vs COVID-19: Book Chapters
0.84
Nilacakra. Srujan, K. S. et al (2018) ‘Classification of Amazon Book Reviews Based on Sentiment Analysis’, in, pp.
Nilacakra Srujan, K. S. et al (2018) ‘Classification of Amazon Book Reviews Based on Sentiment Analysis’, pp。
0.65
401–411. doi: 10.1007/978-981-10-7 512-4_40.
401–411. doi: 10.1007/978-981-10-7 512-4_40。
0.51
Sutoyo, I. (2018) ‘Implementasi Algoritma Decision Tree untuk Klasifikasi Data Peserta Didik’, Pilar Nusa Information System, 14(2), pp.
崇豊 私。 (2018) 'Implementasi Algoritma Decision Tree untuk Klasifikasi Data Peserta Didik', Pilar Nusa Information System, 14(2), pp。
0.57
217–224. doi:
217–224. Doi:
0.63
Journal of Computing and
ジャーナル・オブ・コンピューティングと
0.62
Mandiri : 10.33480/pilar.v14i2 .70.
マンディリ 10.33480/pilar.v14i2 .70。
0.23
Thaha, A. F. (2020) ‘Dampak covid-19 terhadap UMKM di Indonesia’, BRAND Jurnal Ilmiah Manajemen
Thaha, A. F. (2020) ‘Dampak covid-19 terhadap UMKM di Indonesia’, BRAND Jurnal Ilmiah Manajemen
0.99
Pemasaran, 2(1), pp.
Pemasaran, 2(1), pp。
0.71
147–153. Tosepu, R. et al.
147–153. とせぷ、r. et al。
0.62
(2020) ‘Correlation between weather and Covid-19 pandemic in Jakarta, Indonesia’,
(2020年)「インドネシア・ジャカルタにおける天気と新型コロナウイルスのパンデミックの相関」
0.64
Science of The Total Environment, 725, p. 138436. doi: 10.1016/j.scitotenv. 2020.138436.
全環境科学(725, p. 138436. doi: 10.1016/j.scitotenv. 2020.138436)
0.64
Del Vigna, F. et al.
Del Vigna, F. et al.
0.96
(2017) ‘Hate me, hate me not: Hate speech detection on Facebook’, in First Italian
2017年) “Hate me, hate me not: Hate speech detection on Facebook”, First Italian
0.69
Conference on Cybersecurity (ITASEC17), pp.
サイバーセキュリティ会議(ITASEC17)、p。
0.69
86–95. doi: 10.1051/matecconf/20 1712502035.
86–95. 10.1051/matecconf/20 1712502035
0.65
Vo, B.-K. H. and Collier, N. (2013) ‘Twitter Emotion Analysis in Earthquake Situations’, International
Vo, B.-K. H. and Collier, N. (2013) 'Twitter Emotion Analysis in Earthquake situations', International
0.98
Journal of Computational Linguistics and Applications, 4(1), pp.
Journal of Computational Linguistics and Applications, 4(1), pp.
0.79
159–173. Wardani, N. S., Prahutama, A. and Kartikasari, P. (2020) ‘Analisis Sentimen Pemindahan Ibu Kota Negara Dengan Klasifikasi Naïve Bayes Untuk Model Bernoulli dan Multinomial’, Jurnal Gaussian, 9(3), pp.
159–173. Wardani, N. S., Prahutama, A. and Kartikasari, P. (2020) ‘Analisis Sentimen Pemindahan Ibu Kota Negara Dengan Klasifikasi Na've Bayes Untuk Model Bernoulli dan Multinomial’, Jurnal Gaussian, 9(3), pp。
0.80
237–246. doi: 10.14710/j.gauss.v9i 3.27963.
237–246. doi: 10.14710/j.gauss.v9i 3.27963。
0.51
Warner, W. and Hirschberg, J.
Warner, W. and Hirschberg, J.
0.94
(2012) ‘Detecting Hate Speech on the World Wide Web’, in Workshop on
(2012)'detecting hate speech on the world wide web' in workshop on the world wide web'
0.87
Language in Social Media (LSM 2012), pp.
ソーシャルメディアにおける言語(lsm 2012, pp.)。
0.80
19–26. World Health Organisation (WHO) (2020a) Novel Coronavirus (2019-nCoV) Situation Report-22, 11 at: https://www.who.int/ docs/default-source/ coronaviruse/situati on-
19–26. World Health Organisation (WHO) (2020a) New Coronavirus (2019-nCoV) situation Report-22, 11 at: https://www.who.int/ docs/default-source/ coronaviruse/situati on
0.66
February 2020. Available reports/20200211-sit rep-22-ncov.pdf?sfvr sn=fb6d49b1_2.
World Health Organisation (WHO) (2020b) Novel Coronavirus (2019-nCoV), Situation Report-1, 21 https://www.who.int/ docs/default-source/ coronaviruse/situati on-
World Health Organisation (WHO) (2020b) New Coronavirus (2019-nCoV), situation Report-1, 21 https://www.who.int/ docs/default-source/ coronaviruse/situati on
0.61
2020. Available
2020. 利用可能
0.75
January reports/20200121-sit rep-1-2019-ncov.pdf? sfvrsn=20a99c10_4.
at: Yadav, S. et al (2018) ‘Medical sentiment analysis using social media: towards building a patient assisted system’, in Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018).
at: Yadav, S. et al (2018) 'Medical sentiment analysis using social media: to building a patient assisted system' in Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)。
0.83
Zhang, Y. and Haghani, A.
Zhang, Y. and Haghani, A.
0.94
(2015) ‘A gradient boosting method to improve travel time prediction’, doi:
(2015年)「旅行時間予測を改善するための勾配促進法」
0.66
Transportation Research Part C: Emerging Technologies, 10.1016/j.trc.2015.0 2.019.