A Data Sharing and Integration Technology for Heterogeneous Databases

For a long time, there are a large number of heterogeneous databases on the network, and their heterogeneity is manifested in many aspects. With the development of enterprise informatization and e-government, the system database of each department constitutes a real heterogeneous database framework with its independence and autonomy in the network system of many different functional departments. This paper will design information sharing between heterogeneous databases of network database system of many similar functional departments by using XML data model. The solution of data sharing between heterogeneous databases can accelerate the integration of information systems with departments and businesses as the core among enterprises, form a broader and more efficient organic whole, improve the speed of business processing, broaden business coverage, and strengthen cooperation and exchange among enterprises. In addition, heterogeneous database sharing can avoid the waste of data resources caused by the heterogeneity of database, and promote the availability rate of data resources. Due to the advantages of XML data model, the system has good scalability.

A REAL-TIME APPROACH TO TRANSACTION MANAGEMENT IN HETEROGENEOUS DISTRIBUTED REPLICATED DATABASE SYSTEMS

В работе транзакции рассматриваются в виде гетерогенных файлов от пользователей в качестве входных данных и эффективно развертываются в облачной системе хранения данных и в гетерогенных системах баз данных в качестве выходных данных. In this paper, transactions are considered as heterogeneous files from users as input data and are effectively deployed in a cloud storage system and in heterogeneous database systems as output data.

3D Automated Segmentation of Lower Leg Muscles Using Machine Learning on a Heterogeneous Dataset

Quantitative MRI combines non-invasive imaging techniques to reveal alterations in muscle pathophysiology. Creating muscle-specific labels manually is time consuming and requires an experienced examiner. Semi-automatic and fully automatic methods reduce segmentation time significantly. Current machine learning solutions are commonly trained on data from healthy subjects using homogeneous databases with the same image contrast. While yielding high Dice scores (DS), those solutions are not applicable to different image contrasts and acquisitions. Therefore, the aim of our study was to evaluate the feasibility of automatic segmentation of a heterogeneous database. To create a heterogeneous dataset, we pooled lower leg muscle images from different studies with different contrasts and fields-of-view, containing healthy controls and diagnosed patients with various neuromuscular diseases. A second homogenous database with uniform contrasts was created as a subset of the first database. We trained three 3D-convolutional neuronal networks (CNN) on those databases to test performance as compared to manual segmentation. All networks, training on heterogeneous data, were able to predict seven muscles with a minimum average DS of 0.75. U-Net performed best when trained on the heterogeneous dataset (DS: 0.80 ± 0.10, AHD: 0.39 ± 0.35). ResNet and DenseNet yielded higher DS, when trained on a heterogeneous dataset (both DS: 0.86), as compared to a homogeneous dataset (ResNet DS: 0.83, DenseNet DS: 0.76). In conclusion, a CNN trained on a heterogeneous dataset achieves more accurate labels for predicting a heterogeneous database of lower leg muscles than a CNN trained on a homogenous dataset. We propose that a large heterogeneous database is needed, to make automated segmentation feasible for different kinds of image acquisitions.

Ontology-Based Approach of E-government for Interoperability

There is a need for the exchange and sharing of knowledge between the department of government in the e-government system. Therefore, this paper ‘Ontology-Based Approach of E-government’ will discuss the scope of interoperability. With the e-government ontology, there will be proper semantics by using web ontology language (OWL) which helps to give clearer relation and semantics. To have an ontology, knowledge management is important and architecture/framework design is essential. The development of the e-government system is to serve citizens and organizations. However, e-government systems with heterogeneous database and distributed in nature have made difficult to integrate or interoperate. Therefore, developing a knowledge base (KB) is the major task that e-government focuses on. With KB definition and description, it will ensure clarity about e-government services. Knowledge management is important for e-government and the use of ontology is an effective way in semantic technologies. This ontology will enhance the processing of services and data between different departments in government. This kind of ontology will give a common understanding of knowledge and interoperability between the different departments of government. It will also offer effective and efficient value towards the e-government services by which citizens will be benefitted eventually.

A Matching Method of Heterogeneous Database based on SOM and BP Neural Network

In the process of data integration among heterogeneous databases, it is significantly important to analyze the identical attributes and characteristics of the databases. However, the existing main data attribute matching model has the defects of oversize matching space and low matching precision. Therefore, this paper puts forward a heterogeneous data attribute matching model on the basis of fusion of SOM and BP network through analyzing the attribute matching process of heterogeneous databases. This model firstly matches the heterogeneous data attributes in advance by SOM network to determine the centre scope of attribute data to be matched. Secondly, the accurate match will be carried out through BP network of the standard heterogeneous data various attribute center. Finally, the matching result of the relevant actual database shows that this model can effectively reduce the matching space in the case of complex pattern. As for the large-scale data matching, the matching accuracy is relatively high. The average precision is 89.52%, and the average recall rate is 100%.

A Heterogeneous Distributed Computing Model Based on Web Service

Based on the in-depth study of the existing database synchronization model, in order to improve the cross platform ability of the system and facilitate the construction of small and medium-sized enterprise information platform, this paper proposes a heterogeneous distributed computing scheme based on Web service. The scheme uses JMS to realize the message transmission between systems, and uses web service technology to realize cross platform data reading and writing. In the aspect of distributed transaction processing, the two-phase commit protocol is improved to reduce the probability of system deadlock and effectively ensure the consistency of distributed database data. In order to improve the performance of distributed database system, cache technology is introduced, and the way of integrating cache and database transaction processing is proposed, which effectively ensures the validity of cache data. The architecture is oriented to program developers, who can develop efficient and convenient distributed database system on the basis of this architecture. Finally, this architecture is applied to the background management system of mobile express service. The running results show that the architecture can well meet the business requirements of distributed heterogeneous database system synchronization.

T1K+: A Database for Benchmarking Color Texture Classification and Retrieval Methods

In this paper we present T1K+, a very large, heterogeneous database of high-quality texture images acquired under variable conditions. T1K+ contains 1129 classes of textures ranging from natural subjects to food, textile samples, construction materials, etc. T1K+ allows the design of experiments especially aimed at understanding the specific issues related to texture classification and retrieval. To help the exploration of the database, all the 1129 classes are hierarchically organized in 5 thematic categories and 266 sub-categories. To complete our study, we present an evaluation of hand-crafted and learned visual descriptors in supervised texture classification tasks.