Masoud Nosrati

Masoud Nosrati, Ying Cai

Data outsourcing is a cost-effective solution for data owners to tackle issues such as large volumes of data, huge number of users, and intensive computation needed for data analysis. They can simply upload their databases to a cloud and let it perform all management works, including query processing. One problem with this service model is how query issuers can verify the query results they receive are indeed correct. This concern is legitimate because, as a third party, clouds may not be fully trustworthy, and as a large data center, clouds are ideal targets for hackers. There has been significant work on query result verification, but most consider only simple queries where query results can be attained by checking the raw data against the query conditions directly. In this paper, we consider the problem of enabling users to verify the correctness of the results of analytic queries. Unlike simple queries, analytic queries involve ranking functions to score a database, which makes it difficult to build data structures for verification purposes. We propose two approaches, namely one-signature and multi-signature, and show that they work well on three representative types of analytic queries, including top-k, range, and KNN queries, through both analysis and experiments.

Donglai Zhu, Hao Chen, Hengshuai Yao et al.

The problem of action-conditioned image prediction is to predict the expected next frame given the current camera frame the robot observes and an action selected by the robot. We provide the first comparison of two recent popular models, especially for image prediction on cars. Our major finding is that action tiling encoding is the most important factor leading to the remarkable performance of the CDNA model. We present a light-weight model by action tiling encoding which has a single-decoder feedforward architecture same as [action_video_prediction_honglak]. On a real driving dataset, the CDNA model achieves ${0.3986} \times 10^{-3}$ MSE and ${0.9846}$ Structure SIMilarity (SSIM) with a network size of about {\bfseries ${12.6}$ million} parameters. With a small network of fewer than {\bfseries ${1}$ million} parameters, our new model achieves a comparable performance to CDNA at ${0.3613} \times 10^{-3}$ MSE and ${0.9633}$ SSIM. Our model requires less memory, is more computationally efficient and is advantageous to be used inside self-driving vehicles.

Masoud Nosrati

Process modeling is a suitable tool for improving the business processes. Successful process modeling strongly depends on correct requirements engineering. In this paper, we proposed a combination approach for requirements elicitation for developing business models. To do this, BORE (Business-Oriented Requirements Engineering) method is utilized as the base of our work and it is enriched by the important features of the BDD (Business-driven development) method, in order to make the proposed approach appropriate for modeling the more complex processes. As the main result, our method eventuates in exact requirements elicitation that adapts the customers' needs. Also, it let us avoid any rework in the modeling of process. In this paper, we conduct a case study for the paper submission and publication system of a journal. The results of this study not only give a good experience of real world application of proposed approach on a web-based system, also it approves the proficiency of this approach for modeling the complex systems with many sub-processes and complicated relationships.

Masoud Nosrati, Fakhereh Rahimi, Ronak Karimi

In this research, we present a heuristic method for character recognition. For this purpose, a sketch is constructed from the image that contains the character to be recognized. This sketch contains the most important pixels of image that are representatives of original image. These points are the most probable points in pixel-by-pixel matching of image that adapt to target image. Furthermore, a technique called gravity shifting is utilized for taking over the problem of elongation of characters. The consequence of combining sketch and gravity techniques leaded to a language free character recognition method. This method can be implemented independently for real-time uses or in combination of other classifiers as a feature extraction algorithm. Low complexity and acceptable performance are the most impressive features of this method that let it to be simply implemented in mobile and battery-limited computing devices. Results show that in the best case 86% of accuracy is obtained and in the worst case 28% of recognized characters are accurate.