Browsing by Author "Kabir, Md Rayhan"

Now showing 1 - 2 of 2

E-commerce Merchant Fraud Detection using Machine Learning Approach
(Daffodil International University, 2022-06-20) Hasan, Fahim; Mondal, Sourov Kumar; Kabir, Md Rayhan; Al Mamun, Md Abdullah; Hossen, Md. Sagar; Rahman, Nur Salman
At present, e-commerce has become a global phenomenon. With the great achievement of ecommerce, many are cruel Promotional services are also increasing: with the aim of growing sales, spiteful marketers try to improve their target spectators by improving the outcomes of an illegal search using false travel, shopping, etc. In this report, we read about the problem of deception in major commerce platforms. First, we want to list the merchant fraud, the names of those who have previously committed fraud in the business will be marked on the list. And will train machines using machine learning approach. So that, if a merchant id is given in the system, it can detect whether the id is fraud or not. Our lesson here paper is predictable to hut light on the defense in contradiction of e-commerce fraud of active commerce platforms. In this research report, we proposed a machine learning model to analyze and identify merchant fraud. As a machine learning model, we choose the Random forests, decision tree and logistic regression algorithm for our model.
Novel approach to detect hate speech and profanity on online platforms
(BRAC University, 2021-09) Pritha, Barha Meherun; Islam, Samin; Alam, Tabassum; Kabir, Md Rayhan; Sakeef, Nazmus
Hate speech is becoming more prominent and dominant in the virtual world, with the popularity of social media increasing day by day. People nowadays have various online platforms where they can express their hatred and write offensive speech in the safety of their home. They could even spread false rumors and incite hatred out of nothing. Cyberbullies often verbally attack the sentiments of people with different race, nationality, gender, beliefs and political views. They could also target young children and teenagers. It is also important to note that profane language or some sensitive topic may be bothersome when reached in front of young children and teenagers. It has become necessary for modern technology to detect all those profane and hate speeches so that they can be filtered or removed automatically before they can appear in front of young children or hurt the sentiments of targeted people. However, even though it is easy to detect profanities, it could be difficult to detect all the hate speeches which do not have any offensive or sensitive keywords. It is possible to spot all sorts of hate speeches on social media through the application of machine learning, neural networks and natural language processing. In our study, to identify and recognize hate speeches we will use various models and algorithms. Then we will design and implement an algorithm which will be able to detect hate speech and profane language more efficiently.