Detection of Fake Profiles on Online Matrimony Sites

Our paper focuses on the automation of the task in the field of matrimonial website. In our paper we created and demonstrated an idea to automate the task such as verifying profile, validating the documents provided by users and detection of fake profiles. This features are implemented with the help of Artificial Intelligence/Machine Learning algorithms combined together to avoid any intervention of humans in the task and provide a fully automated website. In this paper we represented our idea for detection of fake profiles on online matrimony sites by using machine learning and deep learning. We created a website of matrimony and on login page user has to upload a document and by OCR (optical character reorganization) with the help of this the data is extracted from image document and matches with data given by user. By using machine learning by anomaly detection the user is genuine or not we can find. Due to lack of labelled examples for in-genuine users, we solve the above problem as anomaly detection problem. In this thesis, we use autoencoder which is widely used algorithm for anomaly detection. We capture user’s behavior, pro�le information and edit history to predict him/her as in-genuine or genuine pro�le. We then treat this problem as a reconstruction task using autoencoder which is trained on a set of genuine pro�les features.