Andrea Kulakov

Row Key Designs of NoSQL Database Tables and Their Impact on Write Performance

2016 24th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP), 2016

In several NoSQL database systems, among which is HBase, only one index is available for the tabl... more In several NoSQL database systems, among which is HBase, only one index is available for the tables, which is also the row key and the clustered index. Using other indexes does not come out of the box. As a result, the row key design is the most important thing when designing tables, because an inappropriate design can lead to detrimental consequences on performances and costs. Particular row key designs are suitable for different problems, and in this paper we analyze the performance, characteristics and applicability of each of them. In particular we investigate the effect of using various techniques for modeling row keys: sequences, salting, padding, hashing, and modulo operations. We propose four different designs based on these techniques and we analyze their performance on different HBase clusters when loading HDFS files with various sizes. The experiments show that particular designs consistently outperform others on differently sized clusters in both execution time and even load distribution across nodes.

Review of Automated Weed Control Approaches: An Environmental Impact Perspective

Communications in Computer and Information Science, 2018

восПитание и его ФормЫ в. а. красавин формироваНие патриотичеСкого СоЗНаНия школьНиков На ДокУмеН... more восПитание и его ФормЫ в. а. красавин формироваНие патриотичеСкого СоЗНаНия школьНиков На ДокУмеНтах и материалах великой отечеСтвеННой войНы аннотация. Статья привлекает внимание к актуальным проблемам патриотического воспитания на основании документов и материалов о Великой Отечественной войне. Подчеркивается, что героизм народа на фронте и в тылу является главной причиной Великой Победы. Работа по обнародованию новых документов и материалов о Великой Отечественной войне будет способствовать развитию патриотических чувств подрастающего поколения. Как свидетельствуют проводимые опросы, тема войны вызывает возрастающий интерес общества. По моему мнению, уроки Великой Отечественной войны и всей Второй мировой войны целесообразно изучать в процессе обучения на уроках истории. Как вырастить патриота? Как отобрать школьному учителю документы, которые бы сформировали у ученика чувство гордости за свою страну? Чтобы ответить на эти вопросы, прежде всего, нужно руководствоваться историческим методом. Особенность и новизна этого труда состоит в том, что на его страницах рассматривается такой важный аспект, как место и роль истории войны в патриотическом воспитании школьников. Известно, что проблемы истории Великой Отечественной войны находятся в центре внимания многочисленных дискуссий в средствах массовой информации, а также в научных и учебных учреждениях, в ходе которых высказываются диаметрально противоположные оценки событий, роли государственных и военных деятелей и др. При этом в ходе дискуссии можно встретить утверждение о том, что социалистическая и фашистская идеология равноценны. В статье обстоятельно рассматривается этот вопрос, справедливо обращается на него внимание молодежи.

Download

Improving Activity Recognition Accuracy in Ambient Assisted Living Systems by Automated Feature Engineering

IEEE Access, 2017

Ambient-assisted living (AAL) is promising to become a supplement of the current care models, pro... more Ambient-assisted living (AAL) is promising to become a supplement of the current care models, providing enhanced living experience to people within context-aware homes and smart environments. Activity recognition based on sensory data in AAL systems is an important task because 1) it can be used for estimation of levels of physical activity, 2) it can lead to detecting changes of daily patterns that may indicate an emerging medical condition, or 3) it can be used for detection of accidents and emergencies. To be accepted, AAL systems must be affordable while providing reliable performance. These two factors hugely depend on optimizing the number of utilized sensors and extracting robust features from them. This paper proposes a generic feature engineering method for selecting robust features from a variety of sensors, which can be used for generating reliable classification models. From the originally recorded time series and some newly generated time series [i.e., magnitudes, first derivatives, delta series, and fast Fourier transformation (FFT)-based series], a variety of time and frequency domain features are extracted. Then, using two-phase feature selection, the number of generated features is greatly reduced. Finally, different classification models are trained and evaluated on an independent test set. The proposed method was evaluated on five publicly available data sets, and on all of them, it yielded better accuracy than when using hand-tailored features. The benefits of the proposed systematic feature engineering method are quickly discovering good feature sets for any given task than manually finding ones suitable for a particular task, selecting a small feature set that outperforms manually determined features in both execution time and accuracy, and identification of relevant sensor types and body locations automatically. Ultimately, the proposed method could reduce the cost of AAL systems by facilitating execution of algorithms on devices with limited resources and by using as few sensors as possible.

Download

Weed Segmentation from Grayscale Tobacco Seedling Images

Advances in Intelligent Systems and Computing, 2016

Manual weed extraction from young seedlings is a hard manual labour process. It has to be continu... more Manual weed extraction from young seedlings is a hard manual labour process. It has to be continuously performed to increase the yield per land unit of any agricultural product. Precise segmentation of plant images is an important step towards creating a camera sensor for weed detection. In this paper we present a machine learning approach for segmenting weed parts from images. A dataset has been generated using bumblebee camera under various light conditions and subsequently training and test patches were extracted. We have generated various texture-based descriptors and used different classification algorithms aiming at correctly recognizing weed patches. The results show that in a case when the images are grayscale, the light conditions are varying, and the distance of the camera to the weeds is not constant machine learning algorithms perform poorly.

The impact of the information technology on the business sector development

With the development of information technology, the possibilities for expansion and growth of bus... more With the development of information technology, the possibilities for expansion and growth of businesses are increasing. The interaction between business sector and information technology is bringing other new concepts which are affecting the future of business and information technology as two inseparable sciences. Managing business performance is a critical requirement for maximizing business profitability. In fact, in many situations, it is a critical requirement for remaining feasible in today’s competitive business environment. Company managers are feeling pressure from intolerance for missed performance targets. And with performance measurement periods becoming shorter, management simply must have the capability for rapid reaction. To do this requires monitoring and tracking capabilities that can generate complete and accurate information upon which they can directly react. This information provides the required business intelligence for proactively managing business performance.

Сензорски мрежи за надгледување и контрола на производство на вино

Постојат многу студии за примена на сензорските мрежи во различни области, како полјоделство, сле... more Постојат многу студии за примена на сензорските мрежи во различни области, како полјоделство, следење на животната средина, здравствена заштита, итн. Идеата на овој проект е да се креираат сензорски мрежи за следење на условите во винариите, како и климатските условите при одгледувањето на грозјето. Складираните сензорски податоци може да се анализираат со цел да се открие корелација меѓу надгледуваните параметри од средината и добиениот квалитет на виното. Со примена на техниките за кластерирање, класификација и методите од машинско учење ќе се креира систем кој врз основа на параметрите ќе донесува интелигентни одлуки за одредени активности кои треба да се преземат при одгледувањето на грозјето или при процесот на производство на вино. На пример, податоците од мерењата се важни за донесување на одлуки при одгледувањето на грозјето, како термин за ѓубрење, примена на хемиски средства или наводнување на почвата. Очекуваме дека податоците собрани при користење на системот ќе бидат ко...

Download

Web-based assessment tool for Teachers Digital Competency

PROCEEDINGS OF THE 8th INTERNATIONAL CONFERENCE ON APPLIED INTERNET AND INFORMATION TECHNOLOGIES, 2018

Even though there are several different frameworks and standards for proving the digital competen... more Even though there are several different frameworks and standards for proving the digital competencies expected from teachers, there is a need for a Web-based assessment tools that allow real, dependable and valid assessment of these competencies. This paper addresses the design challenges related to a software solution for selfand peer-assessment of teachers' digital competencies. In this system the competencies are assessed by a teacher his/her-self or by her/his peers using the performance indicators that are based on the competency model for Teachers.

Download

1 Final Report for the CIS Summer Research Grant Efficient Data Management in Wireless Sensor Networks using Artificial Neural Networks

The general problem of data management in Wireless Sensor Networks (WSNs) is to provide efficient... more The general problem of data management in Wireless Sensor Networks (WSNs) is to provide efficient aggregation of different sensor’s data taking into account the problems of the limited energy of the nodes and their unpredictable failures. Generally, this is solved by reducing the communication among nodes. In order to have an efficient data aggregation performance, a preprocessing is needed which would reduce the amount of data being sent over the communication channels. As an outcome of this research, we propose two similar architectures for data aggregation of sound and video signals. These classification architectures have the same core consisted of a modified FuzzyART neural network and a modified SEQUITUR algorithm used previously only for analysis of symbolic sequences. The proposed architectures have been tested in a prototype implementation using Pocket PCs having microphones and cameras as sensors. Furthermore, in the pursuit for efficient data management methods in wireles...

Cloud-based architecture for automated weed control

IEEE EUROCON 2017 -17th International Conference on Smart Technologies, 2017

Automated weed control has received an increased interest from the scientific community in recent... more Automated weed control has received an increased interest from the scientific community in recent years. Albeit there are plenty approaches and even commercially available systems for weed control, several challenges need to be assessed. Most of them use automated detection of weed and apply herbicides with sprayers on the most infested regions of the land. Automated weed control has proven to reduce the quantity of applied herbicides, thus lowering the pollution of products, land and water. Automated weed control can be performed only by accessing large amount of the field sensory data including images and videos from unmanned air and ground vehicles. Due to the increased resolutions of the used vision sensors, there is even larger need of fast and reliable data processing architectures that allow vast volumes of data to be instantly processed. Furthermore, the weed detection includes computer vision algorithms that have high time and space complexity and that often depend on parameters that need to be tuned. By gathering and processing data from multiple fields, the parameter estimation can be performed with higher accuracy and greater reliability. In this paper we propose a cloud-based architecture that improves the automated weed control by leveraging the cloud to gather additional aggregated knowledge from the process of automated weed control and further improve the process of weed control data processing and parameter estimation. We discuss the main benefits of the proposed architecture and the challenges that need to be overcome for it to be accepted by the agricultural communities.

Mobile phone applications for motivating physical activity

The goal of the research described in this article was to determine the possibilities of using th... more The goal of the research described in this article was to determine the possibilities of using the widely spread every day used smartphones to motivate physical activity among the youngest users. Another aim was to find out to which degree these devices and applications can replace the expensive exergaming equipment offered on the market. Three prototype applications were developed and tested with the target group. The applications can be used indoor and outdoor, they offer mental training and the possibility of user generated content which makes them highly customizable to the user's needs and ages. The applications are entertaining, educational and at the same time motivate physical activity. Applications were tested in a kinder-garden with a group of 30 children at the age of 5 and 75 primary school students at the age of 10-11 and they were well accepted by both groups. Teachers and parents present during the testing found the applications very helpful for educational purpos...

Download

Weed Detection Dataset with RGB Images Taken Under Variable Light Conditions

ICT Innovations 2017, 2017

Weed detection from images has received a great interest from scientific communities in recent ye... more Weed detection from images has received a great interest from scientific communities in recent years. However, there are only a few available datasets that can be used for weed detection from unmanned and other ground vehicles and systems. In this paper we present a new dataset (i.e. Carrot-Weed) for weed detection taken under variable light conditions. The dataset contains RGB images from young carrot seedlings taken during the period of February in the area around Negotino, Republic of Macedonia. We performed initial analysis of the dataset and report the initial results, obtained using convolutional neural network architectures.

Automatic Recognition of Emotions from Speech

International Journal of Computational Linguistics Research, 2019

This paper presents an approach to recognition of human emotions from speech. Seven emotions are ... more This paper presents an approach to recognition of human emotions from speech. Seven emotions are recognized: anger, fear, sadness, happiness, boredom, disgust and neutral. The approach is applied on a speech database, which consists of simulated and annotated utterances. First, numerical features are extracted from the sound database by using audio feature extractor. Next, the extracted features are standardized. Then, feature selection methods are used to select the most relevant features. Finally, a classification model is trained to recognize the emotions. Three classification algorithms are tested, with SVM yielding the highest accuracy of 89% and 82% using the 10 fold cross-validation and Leave-One-Speaker-Out techniques, respectively. "Sadness" is the emotion which is recognized with highest accuracy.

Download

Review of Automated Weed Control Approaches: An Environmental Impact Perspective

Communications in Computer and Information Science, 2018

восПитание и его ФормЫ в. а. красавин формироваНие патриотичеСкого СоЗНаНия школьНиков На ДокУмеН... more восПитание и его ФормЫ в. а. красавин формироваНие патриотичеСкого СоЗНаНия школьНиков На ДокУмеНтах и материалах великой отечеСтвеННой войНы аннотация. Статья привлекает внимание к актуальным проблемам патриотического воспитания на основании документов и материалов о Великой Отечественной войне. Подчеркивается, что героизм народа на фронте и в тылу является главной причиной Великой Победы. Работа по обнародованию новых документов и материалов о Великой Отечественной войне будет способствовать развитию патриотических чувств подрастающего поколения. Как свидетельствуют проводимые опросы, тема войны вызывает возрастающий интерес общества. По моему мнению, уроки Великой Отечественной войны и всей Второй мировой войны целесообразно изучать в процессе обучения на уроках истории. Как вырастить патриота? Как отобрать школьному учителю документы, которые бы сформировали у ученика чувство гордости за свою страну? Чтобы ответить на эти вопросы, прежде всего, нужно руководствоваться историческим методом. Особенность и новизна этого труда состоит в том, что на его страницах рассматривается такой важный аспект, как место и роль истории войны в патриотическом воспитании школьников. Известно, что проблемы истории Великой Отечественной войны находятся в центре внимания многочисленных дискуссий в средствах массовой информации, а также в научных и учебных учреждениях, в ходе которых высказываются диаметрально противоположные оценки событий, роли государственных и военных деятелей и др. При этом в ходе дискуссии можно встретить утверждение о том, что социалистическая и фашистская идеология равноценны. В статье обстоятельно рассматривается этот вопрос, справедливо обращается на него внимание молодежи.

Download

Improving Activity Recognition Accuracy in Ambient Assisted Living Systems by Automated Feature Engineering

IEEE Access, 2017

Ambient-assisted living (AAL) is promising to become a supplement of the current care models, pro... more Ambient-assisted living (AAL) is promising to become a supplement of the current care models, providing enhanced living experience to people within context-aware homes and smart environments. Activity recognition based on sensory data in AAL systems is an important task because 1) it can be used for estimation of levels of physical activity, 2) it can lead to detecting changes of daily patterns that may indicate an emerging medical condition, or 3) it can be used for detection of accidents and emergencies. To be accepted, AAL systems must be affordable while providing reliable performance. These two factors hugely depend on optimizing the number of utilized sensors and extracting robust features from them. This paper proposes a generic feature engineering method for selecting robust features from a variety of sensors, which can be used for generating reliable classification models. From the originally recorded time series and some newly generated time series [i.e., magnitudes, first derivatives, delta series, and fast Fourier transformation (FFT)-based series], a variety of time and frequency domain features are extracted. Then, using two-phase feature selection, the number of generated features is greatly reduced. Finally, different classification models are trained and evaluated on an independent test set. The proposed method was evaluated on five publicly available data sets, and on all of them, it yielded better accuracy than when using hand-tailored features. The benefits of the proposed systematic feature engineering method are quickly discovering good feature sets for any given task than manually finding ones suitable for a particular task, selecting a small feature set that outperforms manually determined features in both execution time and accuracy, and identification of relevant sensor types and body locations automatically. Ultimately, the proposed method could reduce the cost of AAL systems by facilitating execution of algorithms on devices with limited resources and by using as few sensors as possible.

Download

Weed Segmentation from Grayscale Tobacco Seedling Images

Advances in Intelligent Systems and Computing, 2016

Manual weed extraction from young seedlings is a hard manual labour process. It has to be continu... more Manual weed extraction from young seedlings is a hard manual labour process. It has to be continuously performed to increase the yield per land unit of any agricultural product. Precise segmentation of plant images is an important step towards creating a camera sensor for weed detection. In this paper we present a machine learning approach for segmenting weed parts from images. A dataset has been generated using bumblebee camera under various light conditions and subsequently training and test patches were extracted. We have generated various texture-based descriptors and used different classification algorithms aiming at correctly recognizing weed patches. The results show that in a case when the images are grayscale, the light conditions are varying, and the distance of the camera to the weeds is not constant machine learning algorithms perform poorly.

Robust histogram-based feature engineering of time series data

Proceedings of the 2015 Federated Conference on Computer Science and Information Systems, 2015

Collecting data at regular time nowadays is ubiquitous. The most widely used type of data that is... more Collecting data at regular time nowadays is ubiquitous. The most widely used type of data that is being collected and analyzed is financial data and sensor readings. Various businesses have realized that financial time series analysis is a powerful analytical tool that can lead to competitive advantages. Likewise, sensor networks generate time series and if they are properly analyzed can give a better understanding of the processes that are being monitored. In this paper we propose a novel generic histogram-based method for feature engineering of time series data. The preprocessing phase consists of several steps: deseansonalyzing the time series data, modeling the speed of change with first derivatives, and finally calculating histograms. By doing all of those steps the goal is threefold: achieve invariance to different factors, good modeling of the data and preform significant feature reduction. This method was applied to the AAIA Data Mining Competition 2015, which was concerned with recognition of activities carried out by firefighters by analyzing body sensor network readings. By doing that we were able to score the third place with predictive accuracy of about 83%, which was about 1% worse than the winning solution.

Download

Transformation of nominal features into numeric in supervised multi-class problems based on the weight of evidence parameter

Proceedings of the 2015 Federated Conference on Computer Science and Information Systems, 2015

Machine learning has received increased interest by both the scientific community and the industr... more Machine learning has received increased interest by both the scientific community and the industry. Most of the machine learning algorithms rely on certain distance metrics that can only be applied to numeric data. This becomes a problem in complex datasets that contain heterogeneous data consisted of numeric and nominal (i.e. categorical) features. Thus the need of transformation from nominal to numeric data. Weight of evidence (WoE) is one of the parameters that can be used for transformation of the nominal features to numeric. In this paper we describe a method that uses WoE to transform the features. Although the applicability of this method is researched to some extent, in this paper we extend its applicability for multi-class problems, which is a novelty. We compared it with the method that generates dummy features. We test both methods on binary and multi-class classification problems with different machine learning algorithms. Our experiments show that the WoE based transformation generates smaller number of features compared to the technique based on generation of dummy features while also improving the classification accuracy, reducing memory complexity and shortening the execution time. Be that as it may, we also point out some of its weaknesses and make some recommendations when to use the method based on dummy features generation instead.

Download

Parallel computation of information gain using Hadoop and MapReduce

Proceedings of the 2015 Federated Conference on Computer Science and Information Systems, 2015

Nowadays, companies collect data at an increasingly high rate to the extent that traditional impl... more Nowadays, companies collect data at an increasingly high rate to the extent that traditional implementation of algorithms cannot cope with it in reasonable time. On the other hand, analysis of the available data is a key to the business success. In a Big Data setting tasks like feature selection, finding discretization thresholds of continuous data, building decision threes, etc are especially difficult. In this paper we discuss how a parallel implementation of the algorithm for computing the information gain can address these issues. Our approach is based on writing Pig Latin scripts that are compiled into MapReduce jobs which then can be executed on Hadoop clusters. In order to implement the algorithm first we define a framework for developing arbitrary algorithms and then we apply it for the task at hand. With intent to analyze the impact of the parallelization, we have processed the FedCSIS AAIA'14 dataset with the proposed implementation of the information gain. During the experiments we evaluate the speedup of the parallelization compared to a one-node cluster. We also analyze how to optimally determine the number of map and reduce tasks for a given cluster. To demonstrate the portability of the implementation we present results using an on-premises and Amazon AWS clusters. Finally, we illustrate the scalability of the implementation by evaluating it on a replicated version of the same dataset which is 80 times larger than the original.

Download

Structure from motion obtained from low quality images in indoor environment

Structure from motion is the process of extracting 3D structure from images taken through the mot... more Structure from motion is the process of extracting 3D structure from images taken through the motion of the camera. The result is dependent on the quality and resolution of the images that are being taken, so it is beneficial to use as high quality images as possible. Sometimes it is not possible to obtain high level of detail in photos because of the environmental, economic or other restrictions. In this paper we analyze structure from motion when using a low resolution camera in indoor environment. The obtained results are compared with the same process when using images of higher resolution and with the 3D structure created from points taken with the Microsoft Kinect sensor.

Structure from motion obtained from low quality images in indoor environment

Structure from motion is the process of extracting 3D structure from images taken through the mot... more Structure from motion is the process of extracting 3D structure from images taken through the motion of the camera. The result is dependent on the quality and resolution of the images that are being taken, so it is beneficial to use as high quality images as possible. Sometimes it is not possible to obtain high level of detail in photos because of the environmental, economic or other restrictions. In this paper we analyze structure from motion when using a low resolution camera in indoor environment. The obtained results are compared with the same process when using images of higher resolution and with the 3D structure created from points taken with the Microsoft Kinect sensor.

Related Authors

Uploads

Papers by Andrea Kulakov

Log In