Search

With the development of Internet technology, the explosive growth of Internet information presentation has led to difficulty in filtering effective information. Finding a model with high accuracy for text classification has become a critical problem to be solved by text filtering, especially for Chinese texts. This paper selected the manually calibrated Douban movie website comment data for research. First, a text filtering model based on the BP neural network has been built; Second, based on the Term Frequency-Inverse Document Frequency (TF-IDF) vector spacemodel and the doc2vec method, the text word frequency vector and the text semantic vector were obtained respectively, and the text word frequency vector was linearly reduced by the Principal Component Analysis (PCA)method. Third, the text word frequency vector after dimensionality reduction and the text semantic vector were combined, add the text value degree, and the text synthesis vector was constructed. Experiments show that the model combined with text word frequency vector degree after dimensionality reduction, text semantic vector, and text value has reached the highest accuracy of 84.67%. [ABSTRACT FROM AUTHOR]

Academic plagiarism has become a serious concern as it leads to the retardation of scientific progress and violation of intellectual property. In this context, we make a study aiming at the detection of cross-linguistic plagiarism based on Natural language Preprocessing (NLP), Embedding Techniques, and Deep Learning. Many systems have been developed to tackle this problem, and many rely on machine learning and deep learning methods. In this paper, we propose Cross-language Plagiarism Detection (CL-PD) method based on Doc2Vec embedding techniques and a Siamese Long Short-Term Memory (SLSTM) model. Embedding techniques help capture the text's contextual meaning and improve the CL-PD system's performance. To show the effectiveness of our method, we conducted a comparative study with other techniques such as GloVe, FastText, BERT, and Sen2Vec on a dataset combining PAN11, JRC-Acquis, Europarl, and Wikipedia. The experiments for the SpanishEnglish language pair show that Doc2Vec+SLSTM achieve the best results compared to other relevant models, with an accuracy of 99.81%, a precision of 99.75%, a recall of 99.88%, an f-score of 99.70%, and a very small loss in the test phase. [ABSTRACT FROM AUTHOR]

Modern deep learning algorithms have achieved tremendous success in many visual applications by training a model with all relevant task-specific data. In this module fake reviews evolving in online political and property data will be discussed. Regarding political information there are many differences while switching from offline to online mode. The views and comments of the politicians in social media act as a question mark "whether it is true or not?". In some cases, it appears to be real and sometimes it may be a false/fake statement. All because, the online information is often considered as the most important source of info for the users these kinds of reviews and comments affects the trust factors of social media. Now days in social media, more of toxic contents and unwanted information were present rather than the useful information. These toxic comments acts as a threat for users using the application. Hence users do not come forward to post their thoughts and actions or to share the information. This module is used to distinguish reviews, identify the factor in online mode of political and property comments in social media. The LSTM and BERT (Bidirectional Encoder Representations from Transformers) algorithms are used in the first module because they offer a wide variety of parameters such as learning rates, input and output biases, and text categorization. Additionally, GPT2 (Generative Pre-Trained Transformer 2) is implemented which helps in text generation, increasing the size of dataset for training in different classification models. Thus it is important to understand for us that ultimately the model will be getting trained in such a way that it is able to give accurate results while we are undergoing classification of text and also while we are undergoing the identification of fake/authentic comments that are there in the source of our data. [ABSTRACT FROM AUTHOR]

Development on Rinca Island by the Indonesian Government has received a lot of reaction from the community. Masses expressed their opinion through social media, especially Twitter regarding the matter. The research was conducted to analyze the public's sentiment about this development which was divided into three categories: pro, contra, and neutral. There are two Doc2Vec models used in this research, the distributed model, and the distributed bag of words, and using support vector machines and logistic regression as classifiers. Each combination of the models and classifier has an accuracy rate above 75% and shows that almost all are against the development of Rinca Island. [ABSTRACT FROM AUTHOR]

Authorship attribution is the task of determining the likely author of a given text, with applications in domains such as literature and literary history, social network analysis, software engineering and cybersecurity. AutoAt , a deep autoencoder-based classification model which exploits the ability of autoencoders to encode meaningful data patterns is proposed to solve this task. Experiments are conducted on a data set of 1571 poems authored by 8 Romanian poets using a distributed document representation. The proposed approach obtains comparable or better results with respect to other machine learning classifiers. Additionally, the formulation of the AutoAt model allows for the computation of the probability that a test instance belongs to a given author class, which may be a useful property in a variety of authorship attribution applications. This aspect and the fact that AutoAt performs well in the difficult task of authorship attribution on poetic data without the step of feature engineering being informed by domain knowledge show that the proposed classifier is a general one, with potential to be used successfully in other fields. [ABSTRACT FROM AUTHOR]

Semantic similarity analysis in natural language texts is getting great attention recently. Semantic analysis of the Quran is especially challenging because it is not simply factual but encodes subtle religious meanings. Investigating similarity and relatedness between the Quranic verses is a hot topic and can promote the acquisition of the underlying knowledge. Therefore, we use an NPL method to detect the semantic-based similarity between the verses of the Quran. The idea is to exploit the distributed representation of text, to learn an informative representation of the Quran's passages. We map the Arabic Quranic verses to numerical vectors that encode the semantic properties of the text. We then measure similarity among those vectors. The performance of our model is judged through cosine similarity between our assigned semantic similarity scores and annotated textual similarity datasets. Our model scored 76% accuracy on detecting the similarity, and it can act as a basis for potential experiments and research. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Information Science & Technology Association/Joho no Kagaku to Gijutsu is the property of Information Science & Technology Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Twitter is one of the major social media platforms today to study human behaviours by analysing their interactions. To ensure popularity of the tweet, the focus should be on the content of the tweet that results in numerous followings of that message with sufficient number of likes and retweets. The high quality of tweets, increases the online reputation of the users who post it. If a user can get the prediction of likes and retweets on his text before posting it on the internet, it would improve the popularity of the tweet from information sharing perspective. In this paper we employed different machine learning classifiers like SVM, Naïve Bayes, Logistic Regression, Random Forest, and Neural Network, on top of two different text processing approaches used in NLP (natural language processing), namely bag-of-words (TFIDF) and word embeddings (Doc2Vec), to check how many likes and retweets can a tweet generate. The results obtained indicate that all the models performed 10-15% better with the bag-of-word technique. [ABSTRACT FROM AUTHOR]

Detecting phishing web pages is a challenging task. The existing detection method for phishing web page based on DOM (Document Object Model) is mainly aiming at obtaining structural characteristics but ignores the overall representation of web pages and the semantic information that HTML tags may have. This paper regards DOMs as a natural language with Doc2Vec model and learns the structural semantics automatically to detect phishing web pages. Firstly, the DOM structure of the obtained web page is parsed to construct the DOM tree, then the Doc2Vec model is used to vectorize the DOM tree, and to measure the semantic similarity in web pages by the distance between different DOM vectors. Finally, the hierarchical clustering method is used to implement clustering of web pages. Experiments show that the method proposed in the paper achieves higher recall and precision for phishing classification, compared to DOM-based structural clustering method and TF-IDF-based semantic clustering method. The result shows that using Paragraph Vector is effective on DOM in a linguistic approach. [ABSTRACT FROM AUTHOR]

Due to the lack of face to face interaction in online learning environment, this article aims essentially to give tutors the opportunity to understand and analyze learners' cognitive behavior. In this perspective, we propose an automatic system to assess learners' cognitive presence regarding their social interactions within asynchronous online discussions. Combining Natural Language Preprocessing, Doc2Vec document embedding method and machine learning techniques; we first make some transformations and pre-processing to the given transcripts, then we apply Doc2Vec method to represent each message as a vector that will be concatenated with LIWC and context features. The vectors are input data of Naïve Bayes algorithm; a machine learning method; that aims to classify transcripts according to cognitive presence categories. [ABSTRACT FROM AUTHOR]

• A novel framework for a TOD index to discover emerging technologies is proposed. • Our framework is proposed using deep learning-based text mining and graph analysis. • The TOD framework reflects information about the ecosystem of new technology-based firms. • The accuracy and validity of our model are demonstrated empirically. • Intensive analysis of emerging technologies derived from the framework is provided. To capture emerging technologies in the fast-changing technology market, use of information concerning new technology-based firms (NTBFs) is strongly encouraged, in addition to the information about the technology itself. Especially, NTBFs rapidly respond to technological change, and their investment information is a significant criterion of technology valuation. Therefore, this study proposes a new technology opportunity discovery (TOD) framework that exploits text mining by deep learning and a knowledge graph (KG) by using three data sources: technology, NTBF, and investor data. First, a technology-classification model was developed using technical text data acquired using Doc2vec and logistic regression, and then this model assigned highly-relevant technology fields to NTBFs using NTBFs' investor relation text data. Next, a KG that considers technology, NTBF, and NTBF's investor was constructed to represent their relations for TOD by using the results of previous steps. Lastly, considering inter-connectivities of such factors, a TOD index that measures the potential of technologies was proposed. The accuracy and validity of the methods were demonstrated empirically, and an evaluation of emerging technologies identified by the analysis was provided. Our framework will be of great significance as a useful alternative to provide new insights for emerging technologies in the industry and market. [ABSTRACT FROM AUTHOR]

• Developing a machine learning approach to measuring the novelty of patents • Integrating doc2vec and the local outlier factor • Representing patents as vectors using textual information of patents • Measuring the novelty of patents on a numerical scale Patent analysis using text mining techniques is an effective way to identify novel technologies. However, the results of previous studies have been of limited use in practice because they require domain-specific knowledge and reflect the limited technological features of patents. As a remedy, this study proposes a machine learning approach to measuring the novelty of patents. At the heart of this approach are doc2vec to represent patents as vectors using textual information of patents and the local outlier factor to measure the novelty of patents on a numerical scale. A case study of 1,877 medical imaging technology patents confirms that our novelty scores are significantly correlated with the relevant patent indicators in the literature and that the novel patents identified have a higher technological impact on average. It is expected that the proposed approach could be useful as a complementary tool to support expert decision-making in identifying new technology opportunities, especially for small and medium-sized companies with limited technological knowledge and resources. [ABSTRACT FROM AUTHOR]

Websites attract millions of visitors due to the convenience of services they offer, which provide for interesting targets for cyber attackers. Most of these websites use JavaScript (JS) to create dynamic content. The exploitation of vulnerabilities in servers, plugins, and other third-party systems enables the insertion of malicious codes into websites. These exploits use methods such as drive-by-downloads, pop up ads, and phishing attacks on news, porn, piracy, torrent or free software websites, among others. Many of the recent cyber-attacks exploit JS vulnerabilities, in some cases employing obfuscation to hide their maliciousness and evade detection. It is, therefore, primal to develop an accurate detection system for malicious JS to protect users from such attacks. This study adopts Abstract Syntax Tree (AST) for code structure representation and a machine learning approach to conduct feature learning called Doc2vec to address this issue. Doc2vec is a neural network model that can learn context information of texts with variable length. This model is a well-suited feature learning method for JS codes, which consist of text content ranging among single line sentences, paragraphs, and full-length documents. Besides, features learned with Doc2Vec are of low dimensions which ensure faster detections. A classifier model judges the maliciousness of a JS code using the learned features. The performance of this approach is evaluated using the D3M dataset (Drive-by-Download Data by Marionette) for malicious JS codes and the JSUNPACK plus Alexa top 100 websites datasets for benign JS codes. We then compare the performance of Doc2Vec on plain JS codes (Plain-JS) and AST form of JS codes (AST-JS) to other feature learning methods. Our experimental results show that the proposed AST features and Doc2Vec for feature learning provide better accuracy and fast classification in malicious JS codes detection compared to conventional approaches and can flag malicious JS codes previously identified as hard-to-detect. [ABSTRACT FROM AUTHOR]