Machine Learning based Private Documents Vault
Ashok Kumar M, Sri Satya Nihanth M, Rajalakshmi Raja, A. Christy
Abstract
With the development of PC frameworks, the amount of sensitive information to be stored as well as the number of threats to this information increases at an unprecedented rate, making information secrecy progressively critical to PC users. At present, for the gadgets generally associated with the Web, the utilization of cloud information capacity services has become practical and normal, permitting fast admittance to such information from any place and at any time. Such easy access undoubtedly carries with it a concern about the secrecy of the information which is conveyed to outsiders. Records may likewise be leaked by the concerned members. A direct arrangement for the client to encode all reports prior to submitting them is essential. This technique, in any case, makes it difficult to proficiently look for archives as they are completely encrypted. This research proposes a confidential record vault with server-side encryption, which also enables clients to easily convey encryption and other security arrangements by offering robust, focused service of encryption keys; moreover, this work creates and implements a model framework. As a result of confirmation, examination, convenience & security, it is demonstrated that the system can fulfil the requirements for information security assurance of sensitive e-records in the open organization environment. Streamlit, a well-known open-source structure is used for implementing AI and perception applications in Python. The initial evaluation shows that this system presents satisfactory results when compared with directly uploading records to a cloud storage service.