Tom 13 (2023): Zeszyt 1 (January 2023)

Outlier detection aims to find a data sample that is significantly different from other data samples. Various outlier detection methods have been proposed and have been shown to be able to detect anomalies in many practical problems. However, in high dimensional data, conventional outlier detection methods often behave unexpectedly due to a phenomenon called the curse of dimensionality. In this paper, we compare and analyze outlier detection performance in various experimental settings, focusing on text data with dimensions typically in the tens of thousands. Experimental setups were simulated to compare the performance of outlier detection methods in unsupervised versus semi-supervised mode and uni-modal versus multi-modal data distributions. The performance of outlier detection methods based on dimension reduction is compared, and a discussion on using k-NN distance in high dimensional data is also provided. Analysis through experimental comparison in various environments can provide insights into the application of outlier detection methods in high dimensional data.

Nowadays, applied computer-oriented and information digitalization technologies are developing very dynamically and are widely used in various industries. One of the highest priority sectors of the economy of Ukraine and other countries around the world, the needs of which require intensive implementation of high-performance information technologies, is agriculture. The purpose of the article is to synthesise scientific and practical provisions to improve the information technology of the comprehensive monitoring and control of microclimate in industrial greenhouses. The object of research is non-stationary processes of aggregation and transformation of measurement data on soil and climatic conditions of the greenhouse microclimate. The subject of research is methods and models of computer-oriented analysis of measurement data on the soil and climatic state of the greenhouse microclimate. The main scientific and practical effect of the article is the development of the theory of intelligent information technologies for monitoring and control of greenhouse microclimate through the development of methods and models of distributed aggregation and intellectualised transformation of measurement data based on fuzzy logic.

Nowadays, textual information grows exponentially on the Internet. Text summarization (TS) plays a crucial role in the massive amount of textual content. Manual TS is time-consuming and impractical in some applications with a huge amount of textual information. Automatic text summarization (ATS) is an essential technology to overcome mentioned challenges. Non-negative matrix factorization (NMF) is a useful tool for extracting semantic contents from textual data. Existing NMF approaches only focus on how factorized matrices should be modeled, and neglect the relationships among sentences. These relationships provide better factorization for TS. This paper suggests a novel non-negative matrix factorization for text summarization (NMFTS). The proposed ATS model puts regularizes on pairwise sentences vectors. A new cost function based on the Frobenius norm is designed, and an algorithm is developed to minimize this function by proposing iterative updating rules. The proposed NMFTS extracts semantic content by reducing the size of documents and mapping the same sentences closely together in the latent topic space. Compared with the basic NMF, the convergence time of the proposed method does not grow. The convergence proof of the NMFTS and empirical results on the benchmark data sets show that the suggested updating rules converge fast and achieve superior results compared to other methods.

Introducing variation in the training dataset through data augmentation has been a popular technique to make Convolutional Neural Networks (CNNs) spatially invariant but leads to increased dataset volume and computation cost. Instead of data augmentation, augmentation of feature maps is proposed to introduce variations in the features extracted by a CNN. To achieve this, a rotation transformer layer called Rotation Invariance Transformer (RiT) is developed, which applies rotation transformation to augment CNN features. The RiT layer can be used to augment output features from any convolution layer within a CNN. However, its maximum effectiveness is shown when placed at the output end of final convolution layer. We test RiT in the application of scale-invariance where we attempt to classify scaled images from benchmark datasets. Our results show promising improvements in the networks ability to be scale invariant whilst keeping the model computation cost low.

Real life applications of deep learning (DL) are often limited by the lack of expert labeled data required to effectively train DL models. Creation of such data usually requires substantial amount of time for manual categorization, which is costly and is considered to be one of the major impediments in development of DL methods in many areas. This work proposes a classification approach which completely removes the need for costly expert labeled data and utilizes noisy web data created by the users who are not subject matter experts. The experiments are performed with two well-known Convolutional Neural Network (CNN) architectures: VGG16 and ResNet50 trained on three randomly collected Instagram-based sets of images from three distinct domains: metropolitan cities, popular food and common objects - the last two sets were compiled by the authors and made freely available to the research community. The dataset containing common objects is a webly counterpart of PascalVOC2007 set. It is demonstrated that despite significant amount of label noise in the training data, application of proposed approach paired with standard training CNN protocol leads to high classification accuracy on representative data in all three above-mentioned domains. Additionally, two straightforward procedures of automatic cleaning of the data, before its use in the training process, are proposed. Apparently, data cleaning does not lead to improvement of results which suggests that the presence of noise in webly data is actually helpful in learning meaningful and robust class representations. Manual inspection of a subset of web-based test data shows that labels assigned to many images are ambiguous even for humans. It is our conclusion that for the datasets and CNN architectures used in this paper, in case of training with webly data, a major factor contributing to the final classification accuracy is representativeness of test data rather than application of data cleaning procedures.