Małgorzata Bach

Vision Object Access and Management with Usage of the In-Memory Databases

Although Poland has been a member of the European Union (EU) for 7 years and internal borders control of the member states have been already lifted (on 21 st December 2007 when Poland joined the Schengen Agreement), the safety of a common external borders problem is still actual. Taking into consideration fact that one of longest segments of EU external land border belong to Poland, it is obvious that our country must look for more effective ways of border protect. Our east border is known for being Union’s back door used by illegal immigrants that often involve in international terrorism. So, as an area of illicit activities, polish border surveillance system has been developed. It is worth noting that currently Poland has participated in very interesting international research project - TALOS, which is aimed at designing, implementing and testing a prototype of adaptable and transportable border inspection system. But such innovative system, applied in place of conventional systems based on static sensors, still copes with former problem of time-consuming data processing. The high speed data acquisition more than previously involves the development of image processing techniques, so the proposals of metadata separation and proposals of replications usage are described in the paper.

Column-Oriented Metadata Organization of Vision Objects

Evolution of an Internet had measurable influence on progress in sphere of vision systems. For 20 years, from the first introduction in the world plug-in camera, this branch of the technology incredible widened one’s possibilities. Nowadays vision systems are widely used in many areas of human life. They are applied, for example, to industrial quality assurance, burglar alarm, intelligent control systems of traffic control, car parks monitoring or to the border crossings control. The activity of such systems is concerned with the need of large amounts of LOB data types storage. In order to satisfy efficient access to such data, it’s necessary to develop adequate mechanisms of data storage and processing. Because of the fact traditional relational systems may be inefficient, the possibility of using the column-oriented database that may speed up the video content management is proposed. Desired efficiency is achieved by images and their metadata separation, and loading the latter (that is: data about data) directly to column-oriented structures. Thus, the search performance of video stored entirely in the single node of the relational databases and with usage of replicas reside in column-oriented databases, was compared.

Document-Oriented Data Stores of Vision Objects

Nowadays vision systems are widely used in many areas of human life. They are applied to industrial quality assurance, burglar alarms, intelligent control systems of traffic control, car parks monitoring or e.g. to the border crossings control. The activity of such systems is concerned with the need of large amounts of LOB data types storage. In order to satisfy efficient access to such data it’s necessary to develop adequate mechanisms of data storage and processing. Because of the fact that traditional relational systems may be inefficient, the possibilities of using NoSQL database that may speed up the video content management are described in the paper.

Cost-Sensitive Feature Selection for Class Imbalance Problem

The class imbalance problem is encountered in real-world applications of machine learning and results in suboptimal performance during data classification. This is especially true when data is not only imbalanced but also high dimensional. The class imbalance is very often accompanied by a high dimensionality of datasets and in such a case these problems should be considered together. Traditional feature selection methods usually assign the same weighting to samples from different classes when the samples are used to evaluate each feature. Therefore, they do not work good enough with imbalanced data. In situation when the costs of misclassification of different classes are diverse, cost-sensitive learning methods are often applied. These methods are usually used in the classification phase, but we propose to take the cost factors into consideration during the feature selection. In this study we analyse whether the use of cost-sensitive feature selection followed by resampling can give good results for mentioned problems. To evaluate tested methods three imbalanced and multidimensional datasets are considered and the performance of chosen feature selection methods and classifiers are analysed.

Performance Aspect of the In-Memory Databases Accessed via JDBC

The conception of storing and managing data directly in RAM appeared some time ago but in spite of very good efficiency, it was impossible to massive implementation because of hardware limitations. Currently, it is possible to store whole databases in memory as well as there are some mechanisms to organize pieces of data as in-memory databases. It has been the interesting issue how this type of databases behaves when accessing via JDBC. Hence we decided to test their performance in terms/sense of the time of SQL query execution. For this purpose TPC Benchmark\(^{\mathrm {TM}}\) H (TPC-H) was applied. In our research we focused on the open source systems such as Altibase, H2, HyperSQL, MariaDB, MySQL Memory.

Project Management in the Scrum Methodology

Taking into account the current leading role of techniques based on the incremental-iterative programming, the system that allows the optimization of project teams developing software in the Agile methodology Scrum technique, is proposed in the paper. Presented tool automates the process of development project management. It is distinguished from the other tools of the same class, by the ability of creation the workflow tasks that allows to solve the problems associated with information and communication in the project team and the kanbanboard visibility that demarcates whether the task is in the process of implementation or in testing. In addition, it implements the algorithm of Sprints’ generating and archiving.

NoSQL E-learning Laboratory—Interactive Querying of MongoDB and CouchDB and Their Conversion to a Relational Database

Nowadays there is a great demand to improve teaching in technical sciences by the intensification of practical skills of students. Realisation of this challenge seems to be especially difficult in context of distance learning, because its tools are generally not equipped with appropriate mechanisms. Our research corresponds with this trend, as we propose to develop Moodle platform with the novel activity modules. In the study we focus on database domain, in particular on the new solutions known as NoSQLs. Two of them, MongoDB and CouchDB datastores, are presented in the paper as well as the implementation of the modules which allow to insert, query, update and delete data in an interactive way is also described. Besides, we prepared the dedicated application in order to map the structures of the given document database into their equivalents of the well-known relational model.

Hybrid Column/Row-Oriented DBMS

The rapid growth of data volumes which store the increasing amount of information makes the necessity of searching for the effective methods of data storing and processing. Some researches on this field recommend changing the row data organization that is classical for DBMS to the columnar one and/or the in-memory approach usage. The article presents chosen hybrid solutions which simultaneously enable storing data both in a row-based way and column-based one, as well as process these data in the in-memory technology.

The Storage Organisation Influence on Database Operations Performance

Nowadays, many database administrators come across problems regarding storing big amount of data. On the one hand there are made demands of improving of data access efficiency but on the other hand, ways of a storage cost reduction are searched for. One of popular approaches to enhancing database systems is using virtualization technologies. They differ in types of storage level access used as well as in expenses necessary for developing an environment. The main purpose of the research presented in the paper was to analyse the influence of an environment used for configuring database system on efficiency of its performance. Four database were taken into consideration: Oracle, MS SQL Server, DB2 and PostgreSQL along with two kinds of storage infrastructures, based on block and file level protocols.