IRJET- THE FAKE NEWS DETECTION USING MACHINE LEARNING ALGORITHMS

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 08 Issue: 10 | Oct 2021 www.irjet.net p-ISSN: 2395-0072 THE FAKE NEWS DETECTION USING MACHINE LEARNING ALGORITHMS Dr. C K GOMATHY1, Ms. C.V.S.VASAVI2, Mr. D.Y.V.RAJESH3, Ms. A.SRIJA4 ------------------------------------------------------------------------***----------------------------------------------------------------------- ABSTRACT: This project comes up with the applications of Logistic Regression: Natural Processing Language (NLP) to detect the true and fake news .These techniques are used for detecting the It is a classification algorithm, used when the fake news happening from some trusted and untrusted value of the target value is categorical in nature .Logistic platforms/non-trustable sources these days. These models Regression is most commonly used when the data in are built using Various Machine Learning techniques to question has binary output .So when it belongs to one class calculate the frequency and count of news from a set. We or another is either 0 or 1. will use several news platforms like twitter and face book for analyzing the news to detect them whether true or Decision Tree Classifier: fake. A Decision Tree is a graphical representation of all Keywords: NLP, Fake News, True News, Logistic possible solutions to a decision based on certain conditions regression, Decision Tree. .On each step or node of a decision tree, used for classification .We try to form a condition on the features to I. INTRODUCTION seperate all the labels or classes contained in a dataset to the fullest purity. Now a day’s fake news is creating different issues from sarcastic articles to fabricated news and government All these classifiers are used to obtain the best accuracy in propaganda in some outlets. Even this fake news was the classification of results. destroying countries and one’s individual peace and harmony and even some families are being destroyed III. EXISTING SYSTEM through this. There exists a large body of research on the topic The data which was collected might contain of machine learning methods for deception detection, most missing values that may lead to inconsistency .To gain of it has been focusing on classifying online reviews and better values the data must be in the media are growing publicly available social media posts, Particularly now a problems with huge ramifications in our society .Obviously days we can’t even trust the news papers. So We used old , a purposely misleading story is a “fake news” but lately classifiers with new modifications. The news taken can be blathering social media ‘s discourse is changing its listed as a small table below. definition. Some of them now use the term to dismiss the Table 1: Representing News Sources facts counter to their preferred view points. II. LITERATURE SURVEY There are several algorithms for detecting fake news. For that we analyze through different classifiers in different way. Some of the classifiers used here are Random Forest Classifier, Logic Regression; Decision Tree Classifiers are the classifiers which we used for obtaining accuracy. Random Forest Classifier: IV. PROPOSED SYSTEM A random forest classifier is a machine learning In the proposed model The existing data models technique that is used to solve regression and classification have been used for training data with few modifications, problems .It utilizes ensemble learning, which is a Later the training models are being tested as multiple technique that combines many classifiers to provide processes like passive aggression model, multinomial solutions to complex problems. A random forest algorithm naive bayes and tested on holdout assets later the trained consists of many decision trees. data is spitted and then generates count victimizer to test and train the data later the results are picked for models. © 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 37  International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 08 Issue: 10 | Oct 2021 www.irjet.net p-ISSN: 2395-0072 Here the diagram representing the training data models. In this way by using these models we can obtain accuracy in results. Fig 2: True News Data Set Fig 1: Representing Training Models IV. METHODOLOGY The methodology for it was how we are using machine learning algorithms in big data for extraction or classification and evaluation of data through various training data models and some random classifiers which are all listed in literature surveys. All these classifiers are used in obtaining the exact Fig 3: Fake News Data Set and nearly values to the problems. It particularly extracts the true and fake news lists in a graphical form and how V. REQUIREMENTS they are impacting all as shown in fig5 in results. The methodology is very simple and quite interesting .All the  Python classifiers implementation can be done by using Jupiter  Numpy notebook tool to gain Results in a Way. Datasets:  Pandas The datasets have been collected from various  Itertools news journals and formed as datasets for two types as true and fake as shown in fig 3&4.All the collected is tested by  Matplotlib using all the classifiers to obtain the accurate results from them. With all these requirements the project is executed in jupyter notebook tool to obtain the accurate results. © 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 38  International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 08 Issue: 10 | Oct 2021 www.irjet.net p-ISSN: 2395-0072 VI. RESULTS After testing the datasets mentioned above with all the classifiers the accurate and approximate results are obtained as follows.fig 5 represents the graphical representation of obtained results.fig 6,7 &8 presents a confusion matrix form of results without normalization. and also the count of real and fake news from datasets. Fig 6: Confusion Matrix, Without Normalization Fig 4: Graph Representing Twitter Fake News 1 Fig 7: Confusion matrix, without normalization VII .CONCLUSION From all these we can conclude that all the results obtained here are accurate of our knowledge and are very easy to understand .As we all know that how people are addicted to social Media now-a-days even for small things Fig 5: Confusion Matrix, Without Normalization they are very dependent on online sources. In that they are forgetting to analyze the thing they have seen all these happens in a large case when it comes to news, So this project is very much helpful in analyzing them to find out the truth. © 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 39  International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 08 Issue: 10 | Oct 2021 www.irjet.net p-ISSN: 2395-0072 VIII. REFERENCES 1) N. J. Conroy, V. L. Rubin, and Y. Chen, “Automatic deception detection: Methods for finding fake news,” Proceedings of the Association for Information Science and Technology, vol. 52, no. 1, pp. 1–4, 2015. 2) S. Feng, R. Banerjee, and Y. Choi, “Syntactic stylometry 2. Mr. D.Y.V.Rajesh, Student, B.E. for deception detection,” in Proceedings of the 50th Annual Computer Science and Engineering, Sri Meeting of the Association for Computational Linguistics: Chandrasekharendra SaraswathiViswa Short Papers-Volume 2, Association for Computational Mahavidyalaya deemed to be university, Enathur, Linguistics, 2012, pp. 171–175. Kanchipuram, India. His Area of Big data analytics 3)Shlok Gilda,Department of Computer Engineering, Evaluating Machine Learning Algorithms for Fake News Detection,2017 IEEE 15th Student Conference on Research and Development (SCOReD) 4. Dr.C K Gomathy, Article: A Semantic Quality of Web Service Information Retrieval Techniques Using Bin Rank A Cloud Monitoring Framework Perform in Web Services, 3. Ms.A.Srija Student, B.E. International Journal of Scientific Research in Computer Computer Science and Engineering, Sri Science Engineering and Information Technology Chandrasekharendra SaraswathiViswa IJSRCSEIT | Volume 3 | Issue 5 | ISSN : 2456-3307,May- Mahavidyalaya deemed to be university, Enathur, 2018 Kanchipuram, India. Her Area of Big data analytics 5.Dr.C K Gomathy, Article: Supply chain-Impact of importance and Technology in Software Release Management, International Journal of Scientific Research in Computer Science Engineering and Information Technology ( IJSRCSEIT ) Volume 3 | Issue 6 | ISSN : 2456- 4. Dr.C.K.Gomathy is Assistant 3307, P.No:1-4, July-2018 Professor in Computer Science and Engineering at 6.Dr.C K Gomathy, Article: A Study on the recent Sri Chandrasekharendra SaraswathiViswa Mahavidyalaya deemed to be university, Enathur, Advancements in Online Surveying , International Journal Kanchipuram, India. Her area of interest is of Emerging technologies and Innovative Research ( JETIR) Volume 5 | Issue 11 | ISSN : 2349-5162, P.No:327-331, Software Engineering, Web Services, Knowledge Management and IOT. Nov-2018 Author’s Profile:- 1. Ms.C.V.S.Vasavi, Student, B.E. Computer Science and Engineering, Sri Chandrasekharendra SaraswathiViswa Mahavidyalaya deemed to be university, Enathur, Kanchipuram, India. Her Area of Big data analytics © 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 40