2019 2nd International Conference on Issues and Challenges in Intelligent Computing Techniques (ICICT)

A SMART SYSTEM FOR FAKE NEWS

DETECTION USING MACHINE LEARNING
Anjali Jain1, Avinash Shakya2, Harsh Khatter3, Amit Kumar Gupta4
1,4
KIET Group of Institutions 2,3ABES Engineering College, Ghaziabad

Abstract: Most of the smart phone users prefer to read the news Fake news detection is made to stop the rumors that are being
via social media over internet. The news websites are publishing spread through the various platforms whether it be social
the news and provide the source of authentication. The question media or messaging platforms, this is done to stop spreading
is how to authenticate the news and articles which are circulated fake news which leads to activities like mob lynching, this has
among social media like WhatsApp groups, Facebook Pages,
been a great reason motivating us to work on this project. We
Twitter and other micro blogs & social networking sites. It is
harmful for the society to believe on the rumors and pretend to have been continuously seeing various news of mob lynching
be a news. The need of an hour is to stop the rumors especially in that leads to the murder of an individual; fake news detection
the developing countries like India, and focus on the correct, works on the objective of detecting this fake news and
authenticated news articles. This paper demonstrates a model stopping activities like this thereby protecting the society from
and the methodology for fake news detection. With the help of these unwanted acts of violence.[1][3][5]
Machine learning and natural language processing, author tried
to aggregate the news and later determine whether the news is The main objective is to detect the fake news, which is a
real or fake using Support Vector Machine. The results of the classic text classification problem with a straight forward
proposed model is compared with existing models. The proposed
proposition. It is needed to build a model that can differentiate
model is working well and defining the correctness of results upto
93.6% of accuracy. between “Real” news and “Fake” news. This leads to
consequences in social networking sites like Facebook,
Keywords: Artificial Intelligence, Fuzzy Logic, Fuzzy Inference, Instagram, microblogging sites like Twitter and instant
Machine Learning, Naive Based Classifier, News, Prediction, messaging applications like WhatsApp, Hike where these fake
Recommendation, Support Vector Machine (SVM). news gets a major boost and gets viral among people, around
the country and globe.[2]The proposed system helps to find the
authenticity of the news. If the news is not real, then the user
is suggested with the relevant news article.
I. INTRODUCTION
II. RELATED WORK
In Today's world, anybody can post the content over the
internet. Unfortunately, counterfeit news gathers a lot of 2.1 Overview
consideration over the web, particularly via web-based There have been quite a several initiatives taken to achieve
networking media. Individuals get misdirected and don't fake news detection:
reconsider before flowing such mis-educational pieces to the - In 2018 three students of Vivekananda Education Society’s
most distant part of the arrangement. Such type of activities Institute of Technology, Mumbai published their research
are not good for the society where some rumors or vague news paper on fake news detection. They wrote in their research
evaporates the negative thought among the people or specific paper, social media age has started in 20th century. Eventually
category of people[1]. As fast the technology is moving, on the the web usage is increasing, the posts are increasing, the
same pace the preventive measures are required to deal with number of articles are increasing. They used various
such activities. Broad communications assuming a gigantic techniques and tool to detect fake news like NLP techniques,
job in impacting the general public and as it is normal, a few machine learning, and artificial intelligence.[5][6][7]
people attempt to exploit it. There are numerous sites which - Facebook and WhatsApp are also working on fake news
give false data. They deliberately attempt to bring out detection as they wrote in an article. They have been working
purposeful publicity, deceptions and falsehood under the for almost one year, and it is currently under the alpha
pretense of being true news. Their basic role is to control the phase.[2]
data that can cause open to have confidence in it. There are - Nguyen Vo student of Ho Chi Minh City University of
loads of case of such sites everywhere throughout the world Technology (HCMUT) Cambodia did his research on fake
.Therefore, counterfeit news influences the brains of the news detection and implemented in 2017. He used Bi-
individuals. As indicated by study Scientist accept that directional GRU with Attention mechanism in his project fake
numerous man-made brainpower calculations can help in news detection; Yang et al. originally proposed this
uncovering the bogus news. mechanism. He also used some Deep learning algorithms and

978-1-7281-1772-0 ©2019 IEEE

tried to implement other deep learning models such that Auto- in the paper by the authors as an existing best suitable
Encoders, GAN, CNN. approach with Naïve Bayes. SVM is best suited for binary
- Samir Bajaj of Stanford University published a research classification. There are various news websites and news
paper on fake news detection. He detects fake news with the blogs which allows to work with RSS feeds and import the
help of NLP perspective and implements some other deep references of the news articles. This will helps us in finding
learning algorithm. He took an authentic data set from Signal the news accuracy.
Media News dataset.

Several approaches have been taken to detect the fake news III. PROPOSED MODEL
after massive widespread fake news in recent times. There are
three types of fake news contributors: social bots, trolls, and There are three main modules in the proposed model, named
cyborg users [3][4].Social Bots says, if a social media account is as
being controlled by a computer algorithm, then it is referred to 3.1Aggregator
as a social bot. The social bot can automatically generate News aggregator sites enable clients to see news and updates
content. Secondly, the trolls are real humans who “aim to from different sources at one advantageous area. They get the
disrupt online communities” in hopes of provoking social information, arrange them in labels/classes, and show it
media users into an emotional response. Other one is, Cyborg. organized appropriately for simpler utilization. There are
Cyborg users are the combination of “automated activities various top most sites like google news, feedly, news360, etc.
with human input.”Humans build accounts and use programs which provide the semi-structured news data. The RSS
to perform activities in social media. For false information Aggregator plugins are available for make the things easy.
detection, there are two categories: Linguistic Cue and Basically, aggregator builds the quality and accuracy of/in the
Network Analysis approaches. The methods generally used to news. Data gathering is the main purpose of any of the news
do such type of works are Naïve Bayes Classifier and Support aggregator. A basic methodology is consistently checking RSS
Vector Machines (SVM). Feeds; Article extraction from various news site and gather
information. Regularly applied systems to discover related
2.2 Facebook Works to Stop Misinformation and False News articles are watchword based methodologies. At the point
Facebook in an article quoted they are working to fight the when all procedure is done, they show important or ongoing
spread of false news in two key areas. First is disrupting news on the page.[15]
economic incentives because of most false news in financially
motivated. Second one is, Building new products to curb the 3.2News Authenticator
spread of false news [6]. Some of the preventive measures New authenticator follows some steps to check whether the
taken by facebook are mentioned here: news is true or false. It will compare news which is given by
 Ranking Improvements: News Feed ranks reduce our side with different websites and various news sources if
the prevalence of false news content. that news is found on any news website then it shows the
 Easier Reporting: Determine what is valuable and given news is true, else it shows there has been no such news
what is not. Stories that are flagged as false by our community in last few days. This can help us from fake news. These days’
than might show up lower in the user feed fake news spread very fast because of social media and the
internet. So, news authenticator helps us to detect either the
2.3 WhatsApp Work for Fake News Detection given news is fake or real.
To stop the spread of misinformation, WhatsApp has
implemented some security measures and also fake news 3.3 News Suggestion /Recommendation System
detection, though these are under alpha phase and are yet to be News suggestion suggests recent news and suggests the news
rolled out to the beta users. WhatsApp testing ‘Suspicious related to the news which the user has given for
Link Detection’ feature: This feature will alert uses by putting authentication. If the news is fake, then this news suggestion
a red label on links that it knows to lead to a fake or gives the related news on that topic. The news suggestion
alternative website/news. Additionally, if a message has been suggests the news based on keywords which you give in your
forwarded from a device more than 25 times, the message news which you wish to authenticate.[4]
could be blocked. [2]
IV. METHODOLOGY
2.4 Outcome
As mentioned in the above section, all top most giants are 4.1 Approach
trying to hide their selves from the rumors and focus should be
on true news and authenticated articles. More or less, the Because of the multi-dimensional nature of fake news, the
approaches follow in the extraction are based on machine recognizing the category of news is not so easy. It is obvious
learning and Natural language processing. The classifiers, that a practical technique must contain a few perspectives to
models and analytical algorithms are required to work hand in precisely handle the issue. This is the reason the proposed
hand for the authentication of news articles. SVM will be used strategy is a mix of Naïve Bayes classifier, Support Vector
Machines, and semantic investigation. The proposed strategy classification purposes. It is based on the idea of finding the
is completely made out of Artificial Intelligence draws near, hyper-plane that best divides the dataset into two classes.
which is basic to precisely order between the genuine or the Hyper-planes are decision boundaries that help the machine
fake, rather than utilizing calculations that can't mirror learning model classify the data or data points. How the
subjective capacities. The three-section strategy is a blend classification of the data point is done using a hyper-planes can
between Machine Learning calculations that subdivide into be seen in figure 4.2 depicted below:[11][12]
managed learning procedures, and characteristic language
preparing techniques.

4.1.1 Naive Bayes

A Naive Bayes classifier is a supervised machine learning
algorithm that uses Bayes’ theorem. The variables that are
used to generate the model are independent of each other. It is
proven that this classifier itself provides pretty good
results.[9][10]
𝑛

𝑃((𝑋|𝐶𝑖 ) = ∏ 𝑃(𝑥𝑘 |𝐶𝑖 ) = 𝑃(𝑥1 |𝐶𝑖 ) × 𝑃(𝑥2 |𝐶𝑖 ) × …

𝑘=1
× 𝑃(𝑥𝑛 |𝐶𝑖 )
The classification is conducted by deriving the maximum
posterior, which is the maximal P(Ci|X) with the above
assumption applying to Bayes theorem. This assumption Figure 4.2 Depiction of hyper-plane dividing the dataset into two classes
greatly reduces the computational cost by only counting the
class distribution. Naive Bayes is popular algorithm which is Also, the benefits of utilizing the SVM strategy are that it will
used to find the accuracy of the news whether its real or fake in general be exceptionally precise and performs incredibly
using multinomial Naïve Bayes. There are number of well on datasets that are semi-structures structured. Moreover,
algorithms that focus on common principle, so it is not the this method is truly adaptable since it tends to be utilized to
only algorithm for training such classifiers. To check if the arrange or even decide numbers. Likewise, support vector
news is fake or real naïve Bayes can be used. machines have the capacity to deal with high dimensional
spaces and will in general be memory proficient.[18]

4.1.2 Support Vector Machine (SVM)

SVM is a good algorithm to extract the binary class based on
the data given to the model. In the proposed model, the work
is to classify the article in two categories either true or false. A 4.2 System Architecture
Support Vector Machine (SVM) is a supervised machine
learning algorithm that can be used for both regression and

Figure 4.3 Flow chart – Classifier Training

Fig 4.1 Use of Naïve Bayes Classifier

 suggested news on that topic which is very useful for any user.
In the future, the efficiency and accuracy of the prototype can
be enhanced to a certain level, and also enhance the user
interface of the proposed model.

REFERENCES

[1] M. Granik and V. Mesyura, “Fake news detection using naive

Bayes classifier,” 2017 IEEE 1st Ukr. Conf. Electr. Comput. Eng. UKRCON
2017 - Proc., pp. 900–903, 2017.
[2] https://indianexpress.com/article/technology/social/whatsapp-
fight-against-fake-news-top-features-to-curb-spread-of-misinformation-
5256782/
[3] A. Martínez-Garcia, S. Morris, M. Tscholl, F. Tracy, and P.
Carmichael, “Case-based learning, pedagogical innovation, and semantic web
technologies,” IEEE Trans. Learn. Technol., vol. 5, no. 2, pp. 104–116, 2012.
[4] P. R. Humanante-Ramos, F. J. Garcia-Penalvo, and M. A. Conde-
Gonzalez, “PLEs in Mobile Contexts: New Ways to Personalize Learning,”
Rev. Iberoam. Tecnol. del Aprendiz., vol. 11, no. 4, pp. 220–226, 2016.
[5] T. Granskogen and J. A. Gulla, “Fake news detection: Network
data from social media used to predict fakes,” CEUR Workshop Proc., vol.
Figure 4.4 Flow chart – Proposed Model
2041, no. 1, pp. 59–66, 2017.
[6] R. V. L, C. Yimin, and C. N. J, “Deception detection for news:
Three types of fakes,” Proc. Assoc. Inf. Sci. Technol., vol. 52, no. 1, pp. 1–4,
2016.
[7] V. Rubin, N. Conroy, Y. Chen, and S. Cornwell, “Fake News or
V. IMPLEMENTATION AND RESULTS
Truth? Using Satirical Cues to Detect Potentially Misleading News,” pp. 7–
17, 2016.
For the implementation purpose, the four existing approaches [8] Z. Jin, J. Cao, Y. Zhang, J. Zhou, and Q. Tian, “Novel Visual and
are considered. The results of mentioned four models are Statistical Image Features for Microblogs News Verification,” IEEE Trans.
Multimed., vol. 19, no. 3, pp. 598–608, 2017.
compared with the proposed model, it is found the accuracy
[9] S. Gilda, “Evaluating machine learning algorithms for fake news
among top 200 results is mentioned in the table 5.1. The detection,” IEEE Student Conf. Res. Dev. Inspiring Technol. Humanit.
demonstration is done using python programming on R studio SCOReD 2017 - Proc., vol. 2018–January, pp. 110–115, 2018.
and some machine learning algorithm. [10] Y. Seo, D. Seo, and C. S. Jeong, “FaNDeR: Fake News Detection
Model Using Media Reliability,” IEEE Reg. 10 Annu. Int. Conf.
Table 5.1 Result Comparison Proceedings/TENCON, vol. 2018–October, no. October, pp. 1834–1838,
2019.
Article Accuracy Implementation Method [11] S. Das Bhattacharjee, A. Talukder, and B. V. Balantrapu, “Active
learning based news veracity detection with feature weighting and deep-
R. V. L, C. Yimin, and C.
76% NLP shallow fusion,” Proc. - 2017 IEEE Int. Conf. Big Data, Big Data 2017, vol.
N. J (2016)
2018–January, pp. 556–565, 2018.
M. Granik and V. Mesyura
74% Naive Bayes [12] S. Helmstetter and H. Paulheim, “Weakly supervised learning for
(2017)
fake news detection on Twitter,” Proc. 2018 IEEE/ACM Int. Conf. Adv. Soc.
Y. Seo, D. Seo, and C. S. Networks Anal. Mining, ASONAM 2018, pp. 274–277, 2018.
86.65% CNN
Jeong (2018) [13] S. B. Parikh, V. Patil, and P. K. Atrey, “On the Origin,
Jain A., Khatter H., Proliferation and Tone of Fake News,” Proc. - 2nd Int. Conf. Multimed. Inf.
93.50% Naive Bayes, SVM, NLP
Shakya A. (2019) Process. Retrieval, MIPR 2019, pp. 135–140, 2019.
[14] A. Dey, R. Z. Rafi, S. Hasan Parash, S. K. Arko, and A.
Chakrabarty, “Fake news pattern recognition using linguistic analysis,” 2018
Jt. 7th Int. Conf. Informatics, Electron. Vis. 2nd Int. Conf. Imaging, Vis.
VI. CONCLUSION
Pattern Recognition, ICIEV-IVPR 2018, pp. 305–309, 2019.
It is significant to find the accuracy of news which is available [15] N. Kim, D. Seo, and C. S. Jeong, “FAMOUS: Fake News
on internet. In the paper, the components for recognizing Fake Detection Model Based on Unified Key Sentence Information,” Proc. IEEE
news are discussed. A mindfulness that not all, the fake news Int. Conf. Softw. Eng. Serv. Sci. ICSESS, vol. 2018–November, pp. 617–620,
2019.
will propagate via web-based networking media. Currently, to
[16] R. L. Vander Wal, V. Bryg, and M. D. Hays, “X-Ray
test out the proposed method of Naïve Bayes classifier, SVM, Photoelectron Spectroscopy (XPS) Applied to Soot & What It Can Do for
and NLP are used. In future, ensuing algorithm may provide You,” Notes, pp. 1–35, 2006.
better results with hybrid approaches for the same purpose [17] M. Gahirwal, “Fake News Detection,” International Journal of
Advance Research, Ideas and Innovations in Technology, vol. 4, no. 1, pp.
fulfilment. The mentioned system detects the fake news on the
817–819, 2018.
based on the models applied. Also it had provided some