Data Representation Techniques Research Articles

Evaluation of the Effectiveness of Sonification for Time-series Data Exploration

Astronomy is a discipline primarily reliant on visual data. However, alternative data representation techniques are being explored, in particular “sonification,” namely, the representation of data into sound. While there is increasing interest in the astronomical community in using sonification in research and educational contexts, its full potential is still to be explored. This study measured the performance of astronomers and nonastronomers to detect a transit-like feature in time-series data (i.e., light curves), which were represented visually or auditorily, adopting different data-to-sound mappings. We also assessed the bias that participants exhibited in the different conditions. We simulated the data of 160 light curves with different signal-to-noise ratios. We represented them as visual plots or auditory streams with different sound parameters to represent brightness: pitch, duration, or the redundant duration and pitch. We asked the participants to identify the presence of transit-like features in these four conditions in a session that included an equal number of stimuli with and without transit-like features. With auditory stimuli, participants detected transits with performances above the chance level. However, visual stimuli led to overall better performances compared to auditory stimuli and astronomers outperformed nonastronomers. Visualisations led to a conservative response bias (reluctance to answer “yes, there is a transit”), whereas sonifications led to more liberal responses (proneness to respond “yes, there is a transit”). Overall, this study contributes to understanding how different representations (visual or auditory) and sound mappings (pitch, duration, and duration and pitch) of time-series data affect detection accuracy and biases.

Improving the use of LLMs in radiology through prompt engineering: from precision prompts to zero-shot learning.

Large language models (LLMs) such as ChatGPT have shown significant potential in radiology. Their effectiveness often depends on prompt engineering, which optimizes the interaction with the chatbot for accurate results. Here, we highlight the critical role of prompt engineering in tailoring the LLMs' responses to specific medical tasks. Using a clinical case, we elucidate different prompting strategies to adapt the LLM ChatGPT using GPT4 to new tasks without additional training of the base model. These approaches range from precision prompts to advanced in-context methods such as few-shot and zero-shot learning. Additionally, the significance of embeddings, which serve as a data representation technique, is discussed. Prompt engineering substantially improved and focused the chatbot's output. Moreover, embedding of specialized knowledge allows for more transparent insight into the model's decision-making and thus enhances trust. Despite certain challenges, prompt engineering plays a pivotal role in harnessing the potential of LLMs for specialized tasks in the medical domain, particularly radiology. As LLMs continue to evolve, techniques like few-shot learning, zero-shot learning, and embedding-based retrieval mechanisms will become indispensable in delivering tailored outputs. · Large language models might impact radiological practice and decision-masking.. · However, implementation and performance are dependent on the assigned task.. · Optimization of prompting strategies can substantially improve model performance.. · Strategies for prompt engineering range from precision prompts to zero-shot learning..

CARMEL: Capturing spatio-temporal correlations via time-series sub-window imaging for home appliance classification

Energy management systems (EMS), as enablers of more efficient energy consumption, monitor and manage appliances to help residents be more energy efficient and thus more frugal. Recent appliance detection and identification techniques for such systems rely on machine learning. However, machine learning solutions for appliance classification on existing low-frequency household metering have not yet been thoroughly investigated. In this paper, we propose CARMEL, a new approach for identifying home appliances from load monitoring in building EMS based on a new data representation technique and a new model that leverages spatio-temporal correlations in the new representation. The proposed data representation technique performs dimensionality expansion of time series that scales linearly, rather than quadratically and, together with the proposed model, outperform the state of the art image transformation models by 5 percentage points. Evaluation on 5 different low-frequency household metering datasets, considering 29 appliances in total, shows that the proposed representation and the corresponding resource-aware deep learning architecture (1) achieve an average weighted F1 score of 0.92 and (2) require only 230 labeled samples and 3x fewer epochs to transfer to new households.

A Novel Fuzzy Unsupervised Feature Learning Approach

The effectiveness of machine learning approaches depends on the quality of the data representation. In fact, some representations may mislead such learning approaches upon concealing relevant explanatory variables. Although feature engineering, that utilizes domain knowledge and/or expert supervision, yields typical data representation techniques, generic unsupervised feature learning represents an even more objective alternative to determine relevant attributes and generate optimal feature spaces. In this paper, we propose a new fuzzy unsupervised feature learning approach (FUL) that automatically derives new features by revealing the intrinsic structure of the data. In fact, FUL exploits the clusters and the associated fuzzy memberships generated by a fuzzy C-means algorithm, and devises new basis functions and their corresponding representation. The experiments results showed that FUL overtakes relevant state of the art approaches. It yielded the highest F1-measure with an improvement of 8%, 11%, 3%, and 4% on Parkinson, Epilepsy, Gait, and breast cancer datasets, respectively.

Hypergraph-Regularized Lp Smooth Nonnegative Matrix Factorization for Data Representation

Nonnegative matrix factorization (NMF) has been shown to be a strong data representation technique, with applications in text mining, pattern recognition, image processing, clustering and other fields. In this paper, we propose a hypergraph-regularized Lp smooth nonnegative matrix factorization (HGSNMF) by incorporating the hypergraph regularization term and the Lp smoothing constraint term into the standard NMF model. The hypergraph regularization term can capture the intrinsic geometry structure of high dimension space data more comprehensively than simple graphs, and the Lp smoothing constraint term may yield a smooth and more accurate solution to the optimization problem. The updating rules are given using multiplicative update techniques, and the convergence of the proposed method is theoretically investigated. The experimental results on five different data sets show that the proposed method has a better clustering effect than the related state-of-the-art methods in the vast majority of cases.

A Review on Deep-Learning-Based Cyberbullying Detection

Bullying is described as an undesirable behavior by others that harms an individual physically, mentally, or socially. Cyberbullying is a virtual form (e.g., textual or image) of bullying or harassment, also known as online bullying. Cyberbullying detection is a pressing need in today’s world, as the prevalence of cyberbullying is continually growing, resulting in mental health issues. Conventional machine learning models were previously used to identify cyberbullying. However, current research demonstrates that deep learning surpasses traditional machine learning algorithms in identifying cyberbullying for several reasons, including handling extensive data, efficiently classifying text and images, extracting features automatically through hidden layers, and many others. This paper reviews the existing surveys and identifies the gaps in those studies. We also present a deep-learning-based defense ecosystem for cyberbullying detection, including data representation techniques and different deep-learning-based models and frameworks. We have critically analyzed the existing DL-based cyberbullying detection techniques and identified their significant contributions and the future research directions they have presented. We have also summarized the datasets being used, including the DL architecture being used and the tasks that are accomplished for each dataset. Finally, several challenges faced by the existing researchers and the open issues to be addressed in the future have been presented.

Exploring a deep learning neural architecture for closed Literature-based discovery

Scientific literature presents a wealth of information yet to be explored. As the number of researchers increase with each passing year and publications are released, this contributes to an era where specialized fields of research are becoming more prevalent. As this trend continues, this further propagates the separation of interdisciplinary publications and makes keeping up to date with literature a laborious task. Literature-based discovery (LBD) aims to mitigate these concerns by promoting information sharing among non-interacting literature while extracting potentially meaningful information. Furthermore, recent advances in neural network architectures and data representation techniques have fueled their respective research communities in achieving state-of-the-art performance in many downstream tasks. However, studies of neural network-based methods for LBD remain to be explored. We introduce and explore a deep learning neural network-based approach for LBD. Additionally, we investigate various approaches to represent terms as concepts and analyze the affect of feature scaling representations into our model. We compare the evaluation performance of our method on five hallmarks of cancer datasets utilized for closed discovery. Our results show the chosen representation as input into our model affects evaluation performance. We found feature scaling our input representations increases evaluation performance and decreases the necessary number of epochs needed to achieve model generalization. We also explore two approaches to represent model output. We found reducing the model’s output to capturing a subset of concepts improved evaluation performance at the cost of model generalizability. We also compare the efficacy of our method on the five hallmarks of cancer datasets to a set of randomly chosen relations between concepts. We found these experiments confirm our method’s suitability for LBD.

A Tensor Method based on Enhanced Tensor Nuclear Norm and Hypergraph Laplacian Regularization for Pan-Cancer Omics Data Analysis.

As a powerful data representation technique, tensor robust principal component analysis (TRPCA) has been widely used for clustering and feature selection tasks. However, it ignores the significant difference in singular values of tensor data and the manifold information contained in different views, thereby causing serious degradation of conventional TRPCA performance. In this paper, a novel tensor method based on enhanced tensor nuclear norm and hypergraph Laplacian regularization (ETHLR) is developed to address the above problem. ETHLR can jointly learn the prior knowledge of singular values and high-order manifold structures in the unified tensor space and the view-specific feature spaces, respectively. Specifically, the enhanced tensor nuclear norm, namely, the weighted tensor Schatten p-norm, is used to shrink the singular values by fully considering the salient difference information of singular values and the complementary information embedded in the tensor space; the hypergraph Laplacian constraint helps encode high-order geometric structures among multiple samples in the nonlinear view-specific feature space. Furthermore, we employ inexact augmented Lagrange multipliers (ALM) to optimize the ETHLR method. Numerous experiments on pan-cancer omics data show that the superiority of ETHLR over several state-of-the-art competitors.

Feature Weighted Non-Negative Matrix Factorization.

Non-negative matrix factorization (NMF) is one of the most popular techniques for data representation and clustering and has been widely used in machine learning and data analysis. NMF concentrates the features of each sample into a vector and approximates it by the linear combination of basis vectors, such that the low-dimensional representations are achieved. However, in real-world applications, the features usually have different importance. To exploit the discriminative features, some methods project the samples into the subspace with a transformation matrix, which disturbs the original feature attributes and neglects the diversity of samples. To alleviate the above problems, we propose the feature weighted NMF (FNMF) in this article. The salient properties of FNMF can be summarized as three-fold: 1) it learns the weights of features adaptively according to their importance; 2) it utilizes multiple feature weighting components to preserve the diversity; and 3) it can be solved efficiently with the suggested optimization algorithm. The performance on synthetic and real-world datasets demonstrates that the proposed method obtains the state-of-the-art performance.

A review on 3D image reconstruction on specific and generic objects

With the emergence of various techniques involved in deep learning the researchers of computer vision tends to focus on the strategies such as object recognition and segmentation of image. This has inclined to accomplish the deep learning techniques in 3D reconstruction of both specific and generic objects. As the space for reconstruction of 3D images either in single or multi view has envisioned the researchers to concentrate on the available technologies used for reconstruction. With the available built in methods and technologies in deep learning, the performance of the proposed methods were reviewed and analyzed using several parameters. As the remaking of 2D images is still in the beginning stage, it is important to study the 3D shape representations, various network architecture, methodologies and approaches behind 3D reconstruction. In this work a review of deep learning methods for single or multiple RGB images of specific and generic object 3D reconstruction was done. Several methods and their importance were also discussed along with the challenges encountered and with further research directions. This paper critically analyzes the various 3D Shaped Representations, 3D Data Network Architectures, Depth Estimation methods, Multi View Representations and the Data Representation Techniques.

An Online Interactive Dashboard to Explore Personal Exposure to Air Pollution

Studies increasingly examine individual exposure to air pollution while accounting for person-specific activity-travel patterns. Supporting policymakers and local communities using the resulting data requires transparent and ethical communication of exposure levels to affected individuals and other stakeholders. This paper asks how an interactive online dashboard might represent individual-level air pollution exposure profiles to different audiences while respecting individuals’ geoprivacy. Using data from 37 Oxford (UK) residents, it shows that heterogeneous individual-level exposure profiles can be shared ethically through different combinations of visualisation method, spatial and temporal resolution of data representation and Geomasking techniques for different dashboard user groups.

Visualization of Multiple Ontology Agro Knowledge Mining Model

Agriculture is an important sector which contributes to 17% of the total GDP of the Indian economy. Soil, crop type, location and season play a major role in agriculture. Quality seed, water, soil, chemical composition, disease prevention are important parameters in Quality crops. Growth in agriculture and modern techniques has given out a new dimension to modern agriculture processes which differ from traditional agriculture. This in turn reduces the workload of farmers and increases productivity. Experienced farmers have great knowledge about farming techniques, crop selection, disease prevention, soil composition and crop management techniques and their composition. Due to less productivity, water, labor and pest management knowledge transfer is not done to the next generation. This system attempts to provide a visualization of knowledge management systems. Data visualization is one of the modern techniques for data representation. Agriculture yield, crop selection, soil composition can be represented in a visualization technique which will help the farmers for better understanding than representing the data in table or text. In this paper, a visualization of an agro knowledge mining approach which extracts knowledge from multiple ontology is proposed.

FSOA‐DNFNet: Incremental indexing and image classification using hybrid optimization‐based deep neuro Fuzzy network

AbstractThe rapid evolution and tremendous growth of internet has provided massive growth of unstructured data that leads to a complexity while retrieving dynamic data effectively. The rapid growth in data volume has imposed many challenging constraints, such as necessity to retrieve data completely even if newly arrived samples are occurred and storing of huge volume of data. This has paved a way for concentrating more on incremental learning that functions on information streams. To speed up retrieval, clustering methods and indexes are utilized and periodic updating of clusters is very substantial because of dynamic nature of databases. Moreover, the standard of clustering techniques purely based on data representation techniques, in which traditional methods faced problems like dimensionality explosion and sparsity. To address such limitations, an effectual strategy is developed for incremental indexing and image classification using proposed Feedback Social Optimization Algorithm (FSOA). The image classification is effectively carried out using Deep neuro fuzzy optimizer and it is trained by employing the proposed FSOA and newly FSOA is derived by the integration of Feedback Artificial Tree (FAT) Algorithm and Social Optimization Algorithm (SOA). Moreover, the proposed FSOA has achieved the maximum clustering accuracy of 93.382, the maximum testing accuracy of 94.4, the maximum sensitivity of 91.892, and the maximum specificity of 96.058.

Dual Optimization for Kolmogorov Model Learning Using Enhanced Gradient Descent

Data representation techniques have made a substantial contribution to advancing data processing and machine learning (ML). Improving predictive power was the focus of previous representation techniques, which unfortunately perform rather poorly on the interpretability in terms of extracting underlying insights of the data. Recently, the Kolmogorov model (KM) was studied, which is an interpretable and predictable representation approach to learning the underlying probabilistic structure of a set of random variables. The existing KM learning algorithms using semi-definite relaxation with randomization (SDRwR) or discrete monotonic optimization (DMO) have, however, limited utility to big data applications because they do not scale well computationally. In this paper, we propose a computationally scalable KM learning algorithm, based on the regularized dual optimization combined with enhanced gradient descent (GD) method. To make our method more scalable to large-dimensional problems, we propose two acceleration schemes, namely, the eigenvalue decomposition (EVD) elimination strategy and an approximate EVD algorithm. Furthermore, a thresholding technique by exploiting the error bound analysis and leveraging the normalized Minkowski $\ell_1$-norm, is provided for the selection of the number of iterations of the approximate EVD algorithm. When applied to big data applications, it is demonstrated that the proposed method can achieve compatible training/prediction performance with significantly reduced computational complexity; roughly two orders of magnitude improvement in terms of the time overhead, compared to the existing KM learning algorithms. Furthermore, it is shown that the accuracy of logical relation mining for interpretability by using the proposed KM learning algorithm exceeds $80\%$.

Developing the business valuation technique for Mergers & Acquisitions

Subject. According to empirical data of consulting firms, the synergistic effect does not arise every time, since about 70 percent of M&A deals happen to be unsuccessful, destroying the enterprise value. Objectives. We analyze whether it is possible to make M&A more effective, and develop the business valuation technique for M&A on the basis of the real options method. Methods. The study involves methods of analysis and synthesis, generalization, induction and deduction, methods of systems analysis, statistical methods, chart and graphical techniques for statistical data representation. Results. We analyzed and determined the most applicable method for options premium valuation – the Monte-Carlo method, since it appraises non-listed companies and facilitates computations. As part of equity valuation, we suggest substituting the rating to the assessment of risk components by analyzing corporate performance indicators. Having analyzed potential deals, we proves the zero correlation between the options premium and the maximum premium, thus making a game formula of the synergistic effect. Conclusions and Relevance. The correlation between shareholders’ premium and options premium is identical to the correlation of the maximum premium and options premium. That is, the correlation does not virtually exist. As a result, we made the synergistic effect formula. The conclusions and recommendations herein can be used by business leaders to financially justify non-organic business growth mechanisms, develop the business strategy subsequently. Appraisers may also rely upon the findings to improve the quality of business value prediction models as part of M&A deals.

Temporal Integration of Text Transcripts and Acoustic Features for Alzheimer's Diagnosis Based on Spontaneous Speech.

Background: Advances in machine learning (ML) technology have opened new avenues for detection and monitoring of cognitive decline. In this study, a multimodal approach to Alzheimer's dementia detection based on the patient's spontaneous speech is presented. This approach was tested on a standard, publicly available Alzheimer's speech dataset for comparability. The data comprise voice samples from 156 participants (1:1 ratio of Alzheimer's to control), matched by age and gender.Materials and Methods: A recently developed Active Data Representation (ADR) technique for voice processing was employed as a framework for fusion of acoustic and textual features at sentence and word level. Temporal aspects of textual features were investigated in conjunction with acoustic features in order to shed light on the temporal interplay between paralinguistic (acoustic) and linguistic (textual) aspects of Alzheimer's speech. Combinations between several configurations of ADR features and more traditional bag-of-n-grams approaches were used in an ensemble of classifiers built and evaluated on a standardised dataset containing recorded speech of scene descriptions and textual transcripts.Results: Employing only semantic bag-of-n-grams features, an accuracy of 89.58% was achieved in distinguishing between Alzheimer's patients and healthy controls. Adding temporal and structural information by combining bag-of-n-grams features with ADR audio/textual features, the accuracy could be improved to 91.67% on the test set. An accuracy of 93.75% was achieved through late fusion of the three best feature configurations, which corresponds to a 4.7% improvement over the best result reported in the literature for this dataset.Conclusion: The proposed combination of ADR audio and textual features is capable of successfully modelling temporal aspects of the data. The machine learning approach toward dementia detection achieves best performance when ADR features are combined with strong semantic bag-of-n-grams features. This combination leads to state-of-the-art performance on the AD classification task.

Three-Dimensional Gaze Projection Heat-Mapping of Outdoor Mobile Eye-Tracking Data

The mobilization of eye-tracking for use outside of the laboratory provides new opportunities for the assessment of pedestrian visual engagement with their surroundings. However, the development of data representation techniques that visualize the dynamics of pedestrian gaze distribution upon the environment they are situated within remains limited. The current study addresses this through highlighting how mobile eye-tracking data, which captures where pedestrian gaze is focused upon buildings along urban street edges, can be mapped as three-dimensional gaze projection heat-maps. This data processing and visualization technique is assessed during the current study along with future opportunities and associated challenges discussed.

Hand Gestures Recognition Using Radar Sensors for Human-Computer-Interaction: A Review

Human–Computer Interfaces (HCI) deals with the study of interface between humans and computers. The use of radar and other RF sensors to develop HCI based on Hand Gesture Recognition (HGR) has gained increasing attention over the past decade. Today, devices have built-in radars for recognizing and categorizing hand movements. In this article, we present the first ever review related to HGR using radar sensors. We review the available techniques for multi-domain hand gestures data representation for different signal processing and deep-learning-based HGR algorithms. We classify the radars used for HGR as pulsed and continuous-wave radars, and both the hardware and the algorithmic details of each category is presented in detail. Quantitative and qualitative analysis of ongoing trends related to radar-based HCI, and available radar hardware and algorithms is also presented. At the end, developed devices and applications based on gesture-recognition through radar are discussed. Limitations, future aspects and research directions related to this field are also discussed.

KM Learning for Millimeter-Wave Beam Alignment and Tracking: Predictability and Interpretability

A data representation technique dubbed Kolmogorov model (KM), has been applied to the beam alignment problem in large-dimensional antenna systems. The previous learning-based beam alignment solely focused on utilizing the predictive power of KM, i.e., the capability of predicting the outcome of random variables that are outside the training set, to reduce the beam training overhead. However, a distinctive feature of KM, namely, the interpretability which enables the capability of extracting additional information hidden inside the data, has not yet been exploited. Moreover, the prohibitively high computational complexity of the existing KM learning algorithm offsets the benefits brought by KM and hampers its application to large-scale problems. In this paper, we propose a joint beam alignment and tracking framework by incorporating the predictability and interpretability of KM. Especially, our proposed scheme enables a novel interpretable beam tracking that reveals insights on relations among the sounded observations to alleviate the beam sounding overhead after the initial beam alignment. To reduce the computational complexity of KM learning, two enhancement approaches, based on discrete monotonic optimization (DMO) and dual optimization, respectively, are proffered. Numerical results demonstrate that the proposed methods can reduce the computational cost of the existing KM learning algorithm by up to three orders of magnitude. Furthermore, the proposed methods show superior beam alignment and tracking performance over other state-of-the-art techniques, notably in the low signal-to-noise ratio (SNR) regime.

Constructing Interlinked Learning Resources for Integrated STEM Education Environment

The integration of STEM disciplines is increasingly advocated because it is one way to make learning more connected and relevant for students. How discipline integration can be achieved are varied, with reference to multidisciplinary, interdisciplinary, and transdisciplinary approaches adding to the debates. In integrated STEM education environment, teachers and students need to analyze and synthesize the learning content themselves through e.g. examining complex problems, using a variety of means and developing their own strategies to solve these problems. During this learning process, an explicit, well-organized and easy-to-reference learning resources will support the learning effectively. This paper focuses on how to organize learning resources for different integrated STEM pedagogical methods and learning tasks. Employing the idea of Web of Data and Linked Open Data, we describe the learning resources by semantic annotations understandable to both human and machines, and interlinking relevant resources according to the learning tasks and pedagogical methods. A user friendly resources searching, accessing, and visualization form in integrated STEM environment is also presented. By analyzing the goal and content of learning, an ontology is designed to represent the body of knowledge in STEM education, so the knowledge point in STEM education can be described by vocabularies in the ontology. These vocabularies serve as semantic tags for learning resources. We attach appropriate semantic tags to every learning resource by employing semantic annotation methods, and the learning resources will be organized and interlinked well according to their content and semantic information. Learners will take advantage of this content-based resources structure and improve the learning performance. Because we adopt standard data representation techniques, such as RDF, OWL and SPARQL etc., it will be convenient to implement and deploy the resources organization architecture in learning environment. We design a comparative experiment between semantic interlinked learning resources and non- semantic organized learning resources. Because of the support of semantic tag, it is easier to perform content-based semantic retrieval on semantic interlinked resources. From the aspects of student self-efficacy and learning achievement goals, we prove the effectiveness of interlinked learning resources organization proposed.