No abstract

Advancement in deep learning requires significantly huge amount of data for training purpose, where protection of individual data plays a key role in data privacy and publication. Recent developments in deep learning demonstarte a huge challenge for traditionally used approch for image anonymization, such as model inversion attack, where adversary repeatedly query the model, inorder to reconstrut the original image from the anonymized image. In order to apply more protection on image anonymization, an approach is presented here to convert the input (raw) image into a new synthetic image by applying optimized noise to the latent space representation (LSR) of the original image. The synthetic image is anonymized by adding well designed noise calculated over the gradient during the learning process, where the resultant image is both realistic and immune to model inversion attack. More presicely, we extend the approach proposed by T. Kim and J. Yang, 2019 by using Deep Convolutional Generative Adversarial Network (DCGAN) in order to make the approach more efficient. Our aim is to improve the efficiency of the model by changing the loss function to achieve optimal privacy in less time and computation. Finally, the proposed approach is demonstrated using a benchmark dataset. The experimental study presents that the proposed method can efficiently convert the input image into another synthetic image which is of high quality as well as immune to model inversion attack.

No abstract

In web search based applications, queries are suggested by users to search and investigate web search engines information requirements regarding user. However the queries submitted by user sometimes might not easily understood by search engines , since queries submitted by user might be short representation and should not precisely characterize users" detailed information requirements. Because numerous uncertain queries might cover an extensive assortment of topic and diverse users might desire to obtain information on based on their diverse query. Since the query based recommendation and assessment of user search goals based on user feedback sessions might improve retrieval results of web search engine for users. Determining the various number of user search goals for same query and it doesn"t automatically discover user search goals based on the user specified query , clustering is performed based on K means clustering. Still K means clustering methods have some major pitfalls such as random selection of initial centroid value and selection of K value; it reduces the grouping results of pseudo documents for user search. In order to shortcoming these issues query based recommendation system and optimization based clustering is proposed in this work. In the proposed work primarily concerns a query based recommendation system to understand user search goals through examining user search engine query logs files. It then creates and suggests the queries with the purpose to cover search goals of user. Second propose a novel Gaussian firefly algorithm (GFA) based clustering method to group similar user pseudo-documents from feedback session which can capably replicate user search goals. Then clustering results of proposed GFA is compared with existing Fuzzy C Means (FCM) and K means clustering methods for web search engines. At end of the clustering method the experimentation results is measured based on the classifier parameters such as Average Precision (AP) Voted Average Precision (VAP) and Classified Average Precision (CAP) to assess the performance accuracy of web search engine results based on the restructured web search results.