Readings in information retrieval.

Anomaly detection in surveillance videos is currently a challenge because of the diversity of possible events. We propose a deep convolutional neural network (CNN) that addresses this problem by learning a correspondence between common object appearances (e.g. pedestrian, background, tree, etc.) and their associated motions. Our model is designed as a combination of a reconstruction network and an image translation model that share the same encoder. The former sub-network determines the most significant structures that appear in video frames and the latter one attempts to associate motion templates to such structures. The training stage is performed using only videos of normal events and the model is then capable to estimate frame-level scores for an unknown input. The experiments on 6 benchmark datasets demonstrate the competitive performance of the proposed approach with respect to state-of-the-art methods.

/pdf/anomaly-detection-in-video-sequence-with-appearance-motion-4bb8vt309t.pdf

Anomaly Detection in Video Sequence With Appearance-Motion Correspondence

For task-scheduling problems in cloud computing, a multi-objective optimization method is proposed here. First, with an aim toward the biodiversity of resources and tasks in cloud computing, we propose a resource cost model that defines the demand of tasks on resources with more details. This model reflects the relationship between the user’s resource costs and the budget costs. A multi-objective optimization scheduling method has been proposed based on this resource cost model. This method considers the makespan and the user’s budget costs as constraints of the optimization problem, achieving multi-objective optimization of both performance and cost. An improved ant colony algorithm has been proposed to solve this problem. Two constraint functions were used to evaluate and provide feedback regarding the performance and budget cost. These two constraint functions made the algorithm adjust the quality of the solution in a timely manner based on feedback in order to achieve the optimal solution. Some simulation experiments were designed to evaluate this method’s performance using four metrics: 1) the makespan; 2) cost; 3) deadline violation rate; and 4) resource utilization. Experimental results show that based on these four metrics, a multi-objective optimization method is better than other similar methods, especially as it increased 56.6% in the best case scenario.

A Multi-Objective Optimization Scheduling Method Based on the Ant Colony Algorithm in Cloud Computing

FIELD: information technology. SUBSTANCE: present invention relates to translation from one language to another. Wireless mobile device service is provided with a service for selective downloading information from the server. The downloaded information includes the architecture of the translation, which has means of translating, independent of language, and at least one language dependent translation database. The translation data base which depends on the language includes translation templates and dictionary translations. It can also load specialised database for the chosen city or cities in the world. Translation between languages is carried out through the application of translation database in the means of translating which depends on the language and translation database, specific to a particular city. For simplification of translation process the means of translation implements a user controlled scheme for replacing terms. EFFECT: invention increases quality and convenience of text translation in a natural language. 24 cl, 5 dwg

Methods and systems for translating from one language to another

在Naive Bayes Classifiers模型中，要求父节点下的子节点（特征变量）之间相对独立，然而在现实世界中，特征与特征之间是非独立的、相关的。提出一种预处理方法，实验结果表明，该方法明显地提高了分类精度。

基于Naive Bayes Classifiers的航空影像纹理分类

A novel GA-ELM model for patient-specific mortality prediction over large-scale lab event data

The emergence of novel techniques for automatic anomaly detection in surveillance videos has significantly reduced the burden of manual processing of large, continuous video streams. However, existing anomaly detection systems suffer from a high false-positive rate and also, are not real-time, which makes them practically redundant. Furthermore, their predefined feature selection techniques limit their application to specific cases. To overcome these shortcomings, a dynamic anomaly detection and localization system is proposed, which uses deep learning to automatically learn relevant features. In this technique, each video is represented as a group of cubic patches for identifying local and global anomalies. A unique sparse denoising autoencoder architecture is used, that significantly reduced the computation time and the number of false positives in frame-level anomaly detection by more than 2.5%. Experimental analysis on two benchmark data sets - UMN dataset and UCSD Pedestrian dataset, show that our algorithm outperforms the state-of-the-art models in terms of false positive rate, while also showing a significant reduction in computation time.

Dynamic video anomaly detection and localization using sparse denoising autoencoders

Sentiment Analysis is one of the prominent research fields in Natural Language Processing because of its widespread real-world applications. Customer preferences, options and experiences can be analyzed through social media, reviews, blogs and other online social networking site data. However, due to increasing informal usage of local languages in social media platforms, multi-lingual or code-mixed data is fast becoming a common occurrence. Mixed code is generated when users use more than a single language in social network comments. Such data presents a significant challenge for applications using sentiment analysis and is yet to be fully explored by researchers. Existing sentiment analysis methods applied to monolingual social data are not suitable for code-mixed data due to the inconsistency in the grammatical structure in these sentences. In this paper, a novel method focused on performing effective sentiment analysis of bilingual sentences written in Hindi and English is proposed, that takes into account linguistic code switching and the grammatical transitions between the two considered languages. Experimental evaluation using real-world, code-mixed datasets obtained from Facebook showed that the proposed approach achieved very good accuracy and was also efficient performance-wise.

Domain-specific sentiment analysis approaches for code-mixed social network data

Cloud computing is currently a very popular computing paradigm as it provides ubiquitous, on-demand access as a service to computing resources via the Internet. In spite of offering marked advantages over the traditional style of computing, there are several issues related to load on the computing system and task scheduling to outperform the computation that need to be effectively solved in order to provide better quality of service to the service consumer. Task scheduling is a crucial research area since it affects the system load and performance; and there will always be scope for optimizing existing scheduling algorithms and propose efficient new task scheduling algorithms. Many task scheduling algorithms to resolve this problem have already been proposed — Particle Swarm Optimization, Ant Colony Optimization, Genetic algorithms, Artificial Bee Algorithm etc. In this paper, we propose a hybrid task scheduling algorithm that is based on combining the plus points of bio-inspired algorithms like Ant Colony Optimization and Artificial Bee Algorithm. We show were the strong points of both these algorithms can be utilized and incorporated in order to optimize task scheduling in the cloud algorithm. It is observed that the proposed algorithm gave an improvement of about 19% when compared to the default FCFS scheduling strategy, 11% better than ABC algorithm and performed 9% better than the conventional ACO based task scheduling.

An hybrid bio-inspired task scheduling algorithm in cloud environment

With the emergence of the Programmable Web paradigm, the World Wide Web is evolving into a Web of Services, where data and services can be effectively reused across applications. Given the wide diversity and scale of published Web services, the problem of service discovery is a big challenge for service-based application development. This is further compounded by the limited availability of intelligent categorization and service management frameworks. In this paper, an approach that extends service similarity analysis by using morphological analysis and machine learning techniques for capturing the functional semantics of real-world Web services for facilitating effective categorization is presented. To capture the functional diversity of the services, different feature vector selection techniques are used to represent a service in vector space, with the aim of finding the optimal set of features. Using these feature vector models, services are classified as per their domain, using ensemble machine learning methods. Experiments were performed to validate the classification accuracy with respect to the various service feature vector models designed, and the results emphasize the effectiveness of the proposed approach.

/pdf/semantics-based-web-service-classification-using-5a0vht5xlv.pdf

S Sowmya Kamath

Papers

A novel GA-ELM model for patient-specific mortality prediction over large-scale lab event data

Dynamic video anomaly detection and localization using sparse denoising autoencoders

Domain-specific sentiment analysis approaches for code-mixed social network data

An hybrid bio-inspired task scheduling algorithm in cloud environment

Semantics-based Web service classification using morphological analysis and ensemble learning techniques