Search

The purpose of the article is to present a new approach to geographic data processing using deep neural networks, which provides more efficient, accurate and automated analysis. The methodology includes the use of deep neural networks for automatic identification of characteristics of geographic objects without prior manual processing. Findings include numerical data in the form of tables and graphs demonstrating the performance improvement of GeoDNN compared to traditional methods. This article presents an improved approach to geographic data processing based on the use of deep neural networks (GeoDNN). The proposed technique provides a more efficient and automated analysis of geodata. GeoDNN uses deep neural networks to automatically identify the characteristics of geographic objects without prior manual processing. This avoids human errors and increases the accuracy of the analysis, and self-learning mechanisms ensure continuous improvement using new data. GeoDNN is characterized by high performance when processing large volumes of geodata due to its optimized structure and parallel computations. The GeoDNN architecture is described in detail, including flowcharts, mathematical formulas, and algorithms. Processes of data preparation, their normalization and augmentation, as well as model training with parameters, optimization methods and loss functions are considered. Originality is the application of deep neural networks to automate the processing of geographic data, which allows to avoid human errors, increases the accuracy of the analysis and ensures constant improvement with the use of new data. The practical value of GeoDNN lies in its ability to automate the analysis of geodata, which significantly reduces processing time and reduces the need for manual labor. This makes the system efficient for use in real-world applications such as automating land parcel analysis. [ABSTRACT FROM AUTHOR]

This paper aims to explore the relationship between deep learning and forecasting within the context of the Sustainable Development Goals (SDGs). The primary objective is to systematically review 38 articles published between 2019 and 2023, following PRISMA guidelines, to understand the current landscape of deep learning forecasting for SDGs. Using data from 2019-2023 allows capturing the latest developments in deep learning forecasting for Sustainable Development Goals (SDGs), while excluding data before 2019 and after 2023 is based on the desire to avoid including potentially less relevant or unpublished research and to maintain focus on the most current and contextually relevant literature. The methodological approach involves analyzing the application of deep learning methods for forecasting within various SDG fields and identifying trends, challenges, and opportunities. The literature review results reveal the popularity of LSTM models, challenges related to data availability, and the interconnected nature of SDGs. Additionally, the study demonstrates that deep learning models enhance forecast accuracy and computational performance, as measured by Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and R-squared (R2). The findings underscore the importance of advanced data preparation techniques and the integration of deep learning with SDGs to improve forecasting outcomes. The novelty of this research lies in its comprehensive overview of the current landscape and its valuable insights for researchers, policymakers, and stakeholders interested in advancing sustainable development goals through deep learning forecasting. Finally, the paper suggests future research directions, including exploring the potential of hybrid forecasting models and investigating the impact of emerging technologies on SDG forecasting methodologies. Innovative methods for imputing missing values in deep learning forecasting models could be further explored to enhance predictive accuracy and robustness. [ABSTRACT FROM AUTHOR]

This research paper focuses on the challenges involved in extracting relevant information from the vast amount of textual data available in digital resources. Specifically, we address the complexities associated with tables in scientific papers, which contain crucial data but can be difficult to understand and summarize. In fact, the data required for training such systems is scarce. To overcome this challenge, we propose the development of a high-quality corpus consisting of summaries, both extractive and abstractive, from table content. To develop such a corpus containing pertinent extractiveabstractive summary pair, we used two approaches, namely, rule-based and template-based. The developed dataset was then validated using various automated and manual metrics. Subsequently, two models, viz., T5 and Seq-to-Seq, were trained on this dataset, to generate abstractive summaries from extractive ones. By using extractive summaries as a starting point, the system can produce new sentences that capture the essence of the tables rather than all the specifics. We have used standard evaluation measures like BLEU, ROUGE-L, Adequacy, and Fluency to evaluate both systems. BLEU and ROUGE-L scores of 58.2 and 0.31 for fine-tuned T5 model and 16.25 and 0.18 for the Seq-to-Seq model suggest that our developed dataset can produce coherent abstractive summaries. [ABSTRACT FROM AUTHOR]

Column Editor's Note.This JLA column posits that academic libraries and their services are dominated by information technologies, and that the success of librarians and professional staff is contingent on their ability to thrive in this technology-rich environment. The column will appear in odd-numbered issues of the journal, and will delve into all aspects of library-related information technologies and knowledge management used to connect users to information resources, including data preparation, discovery, delivery and preservation. Prospective authors are invited to submit articles for this column to the editor atkenning.arlitsch@montana.edu. The Taylor Family Digital Library is the central library opened in 2011 at the University of Calgary dedicated to supporting digital scholarship, creativity, analysis and a supportive learning environment for students. The new building is a technologically advanced converged cultural institution, with mandates to continually evolve in order to meet the needs of students and researchers. The infrastructure to support these mandates required research, collaboration and intense planning, resulting in new construction and technology standards for library renovation and construction projects. This pragmatic article is written for those who will follow in similar footsteps; it provides a roadmap for those embarking on the construction of a new technologically advanced library building. [ABSTRACT FROM PUBLISHER]

This paper presents data mining methods that utilize measurements characterized on the arrangement of allotments of limited sets. Segments are normally connected with object characteristics and significant data mining issue, for example, classification, clustering and data preparation which profit by a logarithmic and geometric investigation of the measurement space of parcels. Allotments are normally connected with object traits and significant data mining issue, for example, classification, clustering and data preparation which profit by an arithmetical and geometric investigation of the measurement space of parcels. The proposed determination model that links the data mining systems and the neural system calculation depends on the air contamination verification data acquired at the air quality monitoring stations in Shijiazhuang. From the start, this model uses data mining innovation to discover the components that influence air quality. In addition, it uses this elementary data to prepare the neural system. Finally, the evaluation test of the determining model is evaluated. The proposed air quality anticipation model developed at present responds well to the problem of down-to-earth application, because it has higher decisive precision. [ABSTRACT FROM AUTHOR]

Highlights • We propose an application of longitudinal data mining. • We propose a full knowledge discovery process on a longitudinal dataset. • We described the data preparation process from Longitudinal Ageing Study dataset. • We developed a semi-supervised strategy to identify short-lived and long-lived profiles. • Results show influences from economic, social and health-related aspects on the classification. Abstract Populational studies of human ageing often generate longitudinal datasets with high dimensionality. In order to discover knowledge in such datasets, the traditional knowledge discovery in database task needs to be adapted. In this article, we present a full knowledge discovery process that was performed on a longitudinal dataset, mentioning the singularities of this process. We investigated the English Longitudinal Study of Ageing's (ELSA's) database, employing both semi-supervised and supervised learning techniques to determine and describe the profiles of individuals annotated with the class labels "short-lived" and "long-lived" who participated in the study. We report on the data preprocessing, the clustering task of finding the best sets of representatives of the profiles of each class, and the use of supervised learning to describe these profiles and perform a longitudinal classification on the dataset to investigate how consistently the unlabelled records would fit into the classes. The results show that several aspects are used to discriminate the individuals between the longevity profiles. Those aspects include economic, social and health-related attributes. The findings have pointed towards a need to further investigate the relationships between the different aspects, especially those related to physical health and wellbeing, and how they affect the lifespan of an individual. Furthermore, our methodology and the adopted procedures can be applied to any other data mining applications for longitudinal studies of ageing. [ABSTRACT FROM AUTHOR]

The article discusses coding information on questionnaires into a form for mechanical tabulation. Two methods exist to code the information. The coder can either code each question on a single questionnaire one after another, or code a single question for every questionnaire. The first is beneficial in that it can help establish trends for each respondent, but can create variations between coders. The second eliminates differences between coders, but breaks up the questionnaire. Four professional coders and four student coders were selected for an experiment to test each method. It found that open-ended questions yielded discrepancies that were, in some cases, very large. These discrepancies were not only between the coders, but also between the performance of a coder at a different time.

Performing longitudinal analysis of socio‐economic change in small‐area spatial units such as census tracts presents several methodological complications and requires significant data preparation. Unit boundaries are revised each census year because of changes in population and delineation methodologies. This limits cross‐year comparison since data are not representative of the same spatial units. To address these problems, we have developed an innovative procedure to reduce error when comparing tract‐level data across census years by apportioning data to the same areal units. This paper describes the methods used to create the Canadian Longitudinal Tract Database. Our procedure is a combination of map‐matching techniques, dasymetric overlays, and population‐weighted areal interpolation. The output is a set of tables with apportionment weights pertaining to pairs of unique boundary identifiers across census years, which can be linked with census data or other data with census identifiers that require longitudinal comparison. Key Messages: Neighbourhood change research is challenged by census boundaries being revised each census year.This paper describes the creation of a longitudinal spatial database of census tracts in Canada, bridging tract‐level data for the 1971–2016 quinquennial censuses to a common set of boundaries.Methodology includes map‐matching, dasymetric overlays, and population‐weighted areal interpolation in order to minimize error when boundaries change over time. [ABSTRACT FROM AUTHOR]

Safety Analyst (SA) is a state-of-the-art software system designed to perform all the steps in the roadway safety management process. To implement the system, agencies must first prepare all the required input data and convert them into a standard format required by the system. However, this data preparation effort is extensive and has become a major obstacle to agency adoption of the system. This article describes Florida's efforts in preparing the input data needed to implement SA. It describes the SA data requirements and discusses several data preparation issues that agencies planning to implement SA are likely to encounter. Specifically, the article describes the methods employed to map Florida's variable codes and site subtypes to the required SA counterparts. The article also describes the approaches taken to collect missing data and to manipulate network data including network segmentation to achieve better results and inclusion of segment direction and geographic coordinates to enable data visualization in SA. The article further describes the development of a data converter tool to automatically convert the input data to standard import files used by SA, and the development of local safety performance functions to achieve better results. These experiences could provide useful lessons learned for agencies that have just started with SA or are contemplating its implementation. [ABSTRACT FROM AUTHOR]

This study developed a comprehensive graphical data processing and modeling system named Visual HEIFLOW (VHF) for integrated surface water-groundwater modeling. Its distinctive features include the following. First, VHF uses a generic multivariable-space-time data cube model, which enables the system to efficiently handle large time-series datasets over a large spatial domain. Second, VHF streamlines the entire modeling procedure from data preparation at the very beginning to visualization and analysis of modeling results in a uniform environment. Third, VHF allows updating the land use input at user-specified time points without manual intervention and therefore allows a direct simulation of the hydrological effects of changing land use. The applicability and versatility of VHF were demonstrated in a case study in the Heihe River Basin, the second largest inland river basin in China. The case study also demonstrated that VFH facilitates process understanding and supports management decision-making. [ABSTRACT FROM AUTHOR]

Data comes in all forms, shapes, sizes and complexities. Stored in files and data sets, SAS ® users know all too well that data can be, and often is, problematic and plagued with a variety of issues. Although today's statistical software programs are extremely powerful, they are typically not designed to overcome poor quality data. This paper describes and recommends a comprehensive data preparation and fuzzy matching process to follow to enable improved statistical modeling. Statistical techniques are also available for comparing the results of the process. Most statistical software users are aware that two or more data files can be joined, or combined, without a problem when the data files have identifiers with unique and reliable values. However, many files do not have unique identifiers, or "keys", and need to be joined using character values, like names or E-mail addresses. To add to the difficulty and confusion, these identifiers might be spelled differently, or use different abbreviation or capitalization protocols. This paper describes a versatile 6-step approach to handling data preparation and fuzzy matching issues for improved statistical modeling. The steps include the identification and understanding of potential matching scenarios; exploring data values and data types; data cleaning and validation; data transformation; traditional merge and join techniques; and an assortment of techniques to successfully merge, join and match less than perfect, or "messy", data by doing phonetic matching using special-purpose character-handling functions like the SOUNDEX algorithm, and the SPEDIS, COMPLEV, and COMPGED fuzzy matching functions. Although the programming techniques described in this paper are illustrated using SAS code, many, if not most, of the techniques can be applied to any software platform that supports character-handling capabilities. [ABSTRACT FROM AUTHOR]

Multicollinearity is a phenomenon in which two or more identified predictor variables in a multiple regression model are co-dependent or highly correlated. The presence of this phenomenon can have a negative impact on the analysis as a whole and can severely limit the conclusions of the research study. This paper reviews and provides examples of the different ways in which multicollinearity can affect a research project, how to detect multicollinearity, and how one can reduce its impact through Ridge Regression. [ABSTRACT FROM AUTHOR]

Data preprocessing is an important step for designing classification model. Normalization is one of the preprocessing techniques used to handle the out-of-bounds attributes. This work develops 14 classification models using different learning algorithms for dynamic selection of normalization technique. This work extracts 12 data complexity measures for 48 datasets drawn from the KEEL dataset repository. Each of these datasets is normalized using min–max and z -score normalization technique. G-mean index is estimated for these normalized datasets using Gaussian Kernel Extreme Learning Machine (KELM) in order to determine the best-suited normalization technique. The data complexity measures along with the best-suited normalization technique are used as an input for developing the aforementioned dynamic models. These models predict the best suitable normalization technique based on the estimated data complexity measures of the dataset. The result shows that the model developed using Gaussian Kernel ELM (KELM) and Support Vector Machine (SVM) give promising results for most of the evaluated classification problems. [ABSTRACT FROM AUTHOR]

Background: Scribes are unlicensed professionals trained in medical data entry. Limited data exist on the impact of scribes on provider time management in the emergency department (ED). Time-motion analysis is a tool utilized in business to capture detailed movements and durations to task completion. It offers a means to categorize how providers allocate their time during a clinical shift.Objective: Evaluate the impact of scribes on how ED providers spend their time.Methods: A prospective observational study was conducted to assess scribe impact on provider time utilization. Four research assistants (RAs) observed attending providers on 24 8-h control shifts (without a scribe), and 24 scribed shifts. RAs observed and categorized provider activity. Providers self-reported after-hours documentation times. Two-sample t-tests were used for normally distributed data, and Wilcoxon rank-sum tests were used for skewed data. All tests were two-sided, and p-values < 0.05 were considered statistically significant.Results: Scribes decreased total documentation time both on shift (mean 55.3 vs. 36.4 min, p < 0.001) and post shift (mean 42.5 vs. 23.3 min, p = 0.038). They did not significantly decrease the amount of time spent reviewing the medical records or placing orders, nor did they have an impact on provider time spent at patients' bedside or time spent discussing patient care with team members.Conclusion: The presence of scribes decreased provider documentation time but did not change the amount of time spent at the bedside or communicating with other team members. Scribes may be a potential strategy to decrease the clerical burden. [ABSTRACT FROM AUTHOR]

Simulating and forecasting river discharge using satellite information is one of the most economical ways of measuring discharge, especially in remote areas where in situ gauges are too expensive to install, maintain, and operate. However, satellite signals are affected by climatic factors, such as clouds, mist, dust storms, and smoke from large forest fires. Hence, a reliable method of preparing and treating space-based signals is necessary. In this paper, a method is introduced to remove inaccurate and out-of-range signals having a low correlation with the in situ discharge measurements, using a combination of data classification and outlier detection procedures. To forecast future river discharge using space-based signals, the signal data set should not contain any gaps. Therefore, we introduced a procedure for the missing signals to be estimated by a model calibrated using the measured discharge. This procedure is illustrated using a case study for the White River, near Boston, MA, USA, that involved using a ddata set of three years’ daily signals of the passive microwave information from the Advanced Microwave Scanning Radiometer for Earth Observing System satellite, which are obtained from the difference between thermal emission of wet and dry land surfaces, and the in situ discharge measurements of the river. The best model was selected from 6200 available models using the Pareto front. The optimum model detected 339 samples as outliers and eliminated them from the data set. Subsequently, the signals were calibrated with the in situ information, and the results indicated the superior accuracy in simulating White River discharge using satellite information. [ABSTRACT FROM AUTHOR]

This paper introduces a problem based classification for mining algorithms. Proposed classification specifies which algorithm is more suitable for the solution of which type of problems. It covers the most spectrum of issues and problems related to data mining starting from data preparation, to actual mining, to result in interpretation and ending with result verification. Result verification is the process of verifying the correctness and consistency of the knowledge generated out of mining. The differences between data and process mining paradigms are highlighted and discussed in this paper. An explanation of well-known algorithms for data and process mining is presented, these algorithms are influential to the research community. [ABSTRACT FROM AUTHOR]

With the rapid growth of online education in recent years, Learning Analytics (LA) has gained increasing attention from researchers and educational institutions as an area which can improve the overall effectiveness of learning experiences. However, the lack of guidelines on what should be taken into consideration during application of LA hinders its full adoption. Therefore, this article investigates the issues that should be considered when approaching the design of LA experiences from the data preparation perspective. The obtained results highlight a validated LA framework of twenty-two designing issues that should be considered by various stakeholders in different contexts as well as a set of guidelines which can enhance designing LA experiences. [ABSTRACT FROM AUTHOR]

Feature selection as one of the key problems of data preprocessing is a hot research topic in pattern recognition, machine learning, and data mining. Evaluating the relevance between features based on information theory is a popular and effective method. However, very little research pays attention to the distinguishing ability of feature, i.e., the degree of a feature distinguishes a given sample with other samples. In this paper, we propose a new feature selection method based on the distinguishing ability of feature. First, we define the concept of maximum-nearest-neighbor, and use this concept to discriminate the nearest neighbors of samples. Then, we present a new measure method for evaluating the quality of feature. Finally, the proposed algorithm is tested on benchmark datasets. Experimental results show that the proposed algorithm can effectively select a discriminative feature subset, and performs as well as or better than other popular feature selection algorithms. [ABSTRACT FROM AUTHOR]

Displaying useful and meaningful information from 3D data is known as volume rendering. Ray casting is one of the most frequently used direct volume rendering methods. It consists of data preparation, sampling, classification, compositing, and shading steps. Normal values are needed for efficient shading. However, 3D volumetric data are discrete and cannot be used directly for shading. Hence, the estimation of normal values, at each voxel on the surface, is needed for realistic shading. In normal estimation, the use of small voxel neighborhoods results in the staircase effect. On the other hand, the use of larger voxel neighborhoods causes loss of details in the final image. In this work, an alternative normal estimation method that uses large voxel neighborhoods is proposed for providing smoother images without losing details. [ABSTRACT FROM AUTHOR]

Household consumption has a significant importance for natural gas wholesale companies. These companies make one-day-ahead forecasting daily. However, there are penalties depending on the error of the estimates. These penalties increase exponentially depending on the error rate. Several studies have been done to develop mathematical models to forecast natural gas consumption and minimize the error rate. However, before mathematical model predictions, a previous step, data preparation, is also important. The data must be prepared correctly before the mathematical model. At this point, prior to the mathematical model, selecting the appropriate data set size has a vital role. In this study, one-day-ahead household natural gas consumption is forecasted for different data sizes. Forecasts have been made for the year 2012. For removing insignificant variables, multiple linear regression (MLR) is applied to all data. In this research, 2 particular scenarios are applied for forecasting. In the first scenario, 2 different data set models are prepared. These sets consist of the data collected 6 weeks before the forecasted day. Daily outcomes are added to the data set and the set is applied in a model called Model A. The other model is depicted based on a sliding window idea having 6 weeks of fixed data size with dynamic data inside (Model W6). For the two models, MLR is applied and error rates are compared. Here, Model A has 7 times higher mean absolute percent error (MAPE) than Model W6. In scenario 2, 6 models are studied and compared for the sliding window approach. The models are named according to the weeks involved (e.g., Model W1, Model W6). MAPEs for Model W3, Model W4, Model W5, and Model W6 are obtained as 11.8%, 6.8%, 7.2%, and 8.1%, respectively. The lowest preday error occurs in the 4-week data model with sliding window approach. [ABSTRACT FROM AUTHOR]

Knowledge discovery in multiple databases offers many opportunities and challenges. We have given a number of motivating points on knowledge discovery in multiple databases. In view of further studies on this aspect, we highlight some domains that generated numerous problems on multiple related databases. Activities related to data preprocessing in a multi-database mining environment are also discussed. Important techniques of mining multiple databases are outlined. Many interesting patterns that originated out of multi-database environments are highlighted. We shall witness more research outcomes and investigations as the number of multi-database domains is on the rise. [ABSTRACT FROM AUTHOR]

Causality and responsibility is an essential tool in the database community for providing intuitive explanations for answers/non-answers to queries. Causality denotes the causes for the answers/non-answers to queries, and responsibility represents the degree of a cause which reflects its influence on the answers/non-answers to queries. In this paper, we study the causality and responsibility problem (CRP) for the non-answers to probabilistic reverse skyline queries (PRSQ). We first formalize CRP on PRSQ, and then, we propose an efficient algorithm termed as CP to compute the causality and responsibility for the non-answers to PRSQ. CP first finds candidate causes, and then, it performs verification to obtain actual causes with their responsibilities, during which several strategies are used to boost efficiency. Further, we explore the CRP for the non-answers to reverse skyline queries. Towards this, we extend CP to identify directly all the actual causes and their responsibilities for a non-answer to reverse skyline queries without additional verification. Extensive experiments using both real and synthetic data sets demonstrate the effectiveness and efficiency of our presented algorithms. [ABSTRACT FROM AUTHOR]

For the analysis of the wet pressing process, which was presented in the first part of this paper, a theoretical approach was chosen. This enabled the pre-definition of three quality-related priorities which now will be considered in detail in the second part. For further analysis, real process data, recorded in an early phase of the process implementation, are used. The challenge is that in this process status, the availability of data is limited or the data sets are incomplete. Supported by the theoretical approach, an easier interpretation of the process data, and in case of ambiguous issues, an accelerated decision making is expected. The objective is to show that this combination is suitable for the process analysis in an early production phase. [ABSTRACT FROM AUTHOR]

The decreasing cost of performing genome-wide association studies has made genomics widely accessible. However, there is a paucity of guidance for best practice in conducting such analyses. For the results of a study to be valid and replicable, multiple biases must be addressed in the course of data preparation and analysis. In addition, standardizing methods across small, independent studies would increase comparability and the potential for effective meta-analysis. This article provides a discussion of important aspects of quality control, imputation and analysis of genome-wide data from a lowcoverage microarray, as well as a straight-forward guide to performing a genome-wide association study. A detailed protocol is provided online, with example scripts available at https://github.com/JoniColeman/gwas_scripts. [ABSTRACT FROM AUTHOR]

As processor design has been transiting from homogeneous multicore processor to heterogeneous multicore processor, traditional Amdahl's law cannot meet the new challenges for asymmetric multicore system. In order to further investigate the impact factors related to the Overhead of Data Preparation (ODP) for Asymmetric multicore systems, we evaluate an asymmetric multicore system built with CPU-GPU by measuring the overheads of memory transfer, computing kernel, cache missing and synchronization. This paper demonstrates that decreasing the overhead of data preparation is a promising approach to improve the whole performance of heterogeneous system. [ABSTRACT FROM AUTHOR]

Quantitative analysis of organismal form is an important component for almost every branch of biology. Although generally considered an easily-measurable structure, the quantification of gastropod shell form is still a challenge because many shells lack homologous structures and have a spiral form that is difficult to capture with linear measurements. In view of this, we adopt the idea of theoretical modelling of shell form, in which the shell form is the product of aperture ontogeny profiles in terms of aperture growth trajectory that is quantified as curvature and torsion, and of aperture form that is represented by size and shape. We develop a workflow for the analysis of shell forms based on the aperture ontogeny profile, starting from the procedure of data preparation (retopologising the shell model), via data acquisition (calculation of aperture growth trajectory, aperture form and ontogeny axis), and data presentation (qualitative comparison between shell forms) and ending with data analysis (quantitative comparison between shell forms). We evaluate our methods on representative shells of the genera Opisthostoma and Plectostoma, which exhibit great variability in shell form. The outcome suggests that our method is a robust, reproducible, and versatile approach for the analysis of shell form. Finally, we propose several potential applications of our methods in functional morphology, theoretical modelling, taxonomy, and evolutionary biology. [ABSTRACT FROM AUTHOR]

We describe an introductory data science course, entitled Introduction to Data Science, offered at the University of Illinois at Urbana-Champaign. The course introduced general programming concepts by using the Python programming language with an emphasis on data preparation, processing, and presentation. The course had no prerequisites, and students were not expected to have any programming experience. This introductory course was designed to cover a wide range of topics, from the nature of data, to storage, to visualization, to probability and statistical analysis, to cloud and high performance computing, without becoming overly focused on any one subject. We conclude this article with a discussion of lessons learned and our plans to develop new data science courses. [ABSTRACT FROM AUTHOR]

Artificial Neural Network which possessing self-learning ability has shown great promise in addressing problem of learning an artist style in generating facial caricatures. This paper presents Artificial Neural Network model to imitate a particular artist style and to predict a facial caricature configuration for a given original face image (photo). This paper also describes the data preparing process that proposes a modified procrustes superimposition method in deriving the datasets for the neural network model. The experiment is carried out to compare the modified procrustes superimposition method with the original one and to find the appropriate neural network structure that would yield the most accurate prediction results. Different datasets (N1, N2, and N3) derived from the same raw data but using the different method in preparing the dataset and different numbers of hidden nodes (6, 12, 18 and 24) are tested. The experimental result and its detail analysis are given and discussed. It proves that neural network has an ability to predict how the artist exaggerates the original facial feature point. Dataset N2 which use Modified Procrustes Superimposition method and the simple structure of a single hidden layer neural network, in which 6 is the number of hidden node, give the best accuracy of the prediction. [ABSTRACT FROM AUTHOR]

Data mining techniques are the result of a long process of research and product development. Large amount of data are searched by the practice of Data Mining to find out the trends and patterns that go beyond simple analysis. For segmentation of data and also to evaluate the possibility of future events, complex mathematical algorithms are used here. Specific algorithm produces each Data Mining model. More than one algorithms are used to solve in best way by some Data Mining problems. Data Mining technologies can be used through Oracle. Generalized Linear Models (GLM) Algorithm is used in Regression and Classification Oracle Data Mining functions. For linear modelling, GLM is one the popular statistical techniques. For regression and binary classification, GLM is implemented by Oracle Data Mining. Row diagnostics as well as model statistics and extensive coefficient statistics are provided by GLM. It also supports confidence bounds.. This paper outlines and produces analysis of GLM algorithm, which will guide to understand the tuning, diagnostics & data preparation process and the importance of Regression & Classification supervised Oracle Data Mining functions and it is utilized in marketing, time series prediction, financial forecasting, overall business planning, trend analysis, environmental modelling, biomedical and drug response modelling, etc. [ABSTRACT FROM AUTHOR]

The Digital Repository of Ireland (DRI) is Ireland’s national trusted digital repository for the social and cultural, historical and contemporary data held by Irish institutions. DRI provides users with a bilingual (Irish and English) user interface at all user access levels, and provides innovative ways to process and display bilingual metadata. This article details our experience in enriching the bilingual metadata and developing the bilingual features of the repository. We present solutions to some of the linguistic and technical challenges we faced and provide recommendations to developers and archivists on how best to prepare bilingual content for contemporary archival repositories. [ABSTRACT FROM AUTHOR]

When building models, it is common to split the whole dataset into a development and a validation sample. In some cases, using random sampling instead of stratified sampling can lead to loss of representativeness of final samples. In such cases, a model built on these data gives different or unexpected results when its performance is measured on the validation sample. In the business area, a lack of representativeness can cause interpretative problems and can have a huge financial impact when a biased model is involved in the credit granting process. The aim of this paper is to examine and understand why representativeness should be checked before the start of modelling. The paper deals with methods of identification of selection bias in time. It recommends using three tests as a common part of the data preparation process. [ABSTRACT FROM AUTHOR]

Learning to rank has become increasingly important for many information retrieval applications. To reduce the labeling cost at training data preparation, many active sampling algorithms have been proposed. In this article, we propose a novel active learning-for-ranking strategy called ranking-based sensitivity sampling (RSS), which is tailored for Gradient Boosting Decision Tree (GBDT), a machine-learned ranking method widely used in practice by major commercial search engines for ranking. We leverage the property of GBDT that samples close to the decision boundary tend to be sensitive to perturbations and design the active learning strategy accordingly. We further theoretically analyze the proposed strategy by exploring the connection between the sensitivity used for sample selection and model regularization to provide a potentially theoretical guarantee w.r.t. the generalization capability. Considering that the performance metrics of ranking overweight the top-ranked items, item rank is incorporated into the selection function. In addition, we generalize the proposed technique to several other base learners to show its potential applicability in a wide variety of applications. Substantial experimental results on both the benchmark dataset and a real-world dataset have demonstrated that our proposed active learning strategy is highly effective in selecting the most informative examples. [ABSTRACT FROM AUTHOR]

The process of knowledge discovery in data consists of five steps. Data preparation, which includes data cleaning and feature selection, takes away from 60% to 95% total time of the whole process. Thus, it is crucial phase of the process. The purpose of this research is to investigate feature selection techniques performance by conducting empirical research. Our comparison of three feature selection techniques reveals significant difference in feature selection techniques performance. [ABSTRACT FROM AUTHOR]

Image registration of biological data is challenging as complex deformation problems are common. Possible deformation effects can be caused in individual data preparation processes, involving morphological deformations, stain variations, stain artifacts, rotation, translation, and missing tissues. The combining deformation effects tend to make existing automatic registration methods perform poor. In our experiments on serial histopathological images, the six state of the art image registration techniques, including TrakEM2, SURF + affine transformation, UnwarpJ, bUnwarpJ, CLAHE + bUnwarpJ and BrainAligner, achieve no greater than 70% averaged accuracies, while the proposed method achieves 91.49% averaged accuracy. The proposed method has also been demonstrated to be significantly better in alignment of laser scanning microscope brain images and serial ssTEM images than the benchmark automatic approaches (p < 0.001). The contribution of this study is to introduce a fully automatic, robust and fast image registration method for 2D image registration. [ABSTRACT FROM AUTHOR]

Abstract: Without data preparation, data mining algorithms cannot operate on data within the knowledge discovery in databases (KDD) process. In fact, the success of later KDD phases largely depends on the data preparation stage. The use of mechanisms for automatically preparing data saves a lot of time and resources within the KDD process. These resources will then be available for use at later, less automatable stages, for example, during results interpretation. We have proposed a general-purpose mechanism applicable to multiple domains in order to improve the data preparation phase in the KDD process. This mechanism processes and automatically converts input data to a suitable format for the application of different data preparation techniques based on a known syntax. It is based on the use of description logic. Taking a generic UML2 data model as a reference, this mechanism is able to check whether any XML data source whatsoever can be transformed and modelled as a subsumption or instance of the above UML2 model. Thus it automatically identifies a consistent, non-ambiguous and finite set of XLST transformations which are used to prepare the data for the application of data mining techniques, obviating the need to expend resources on the preliminary preparation and formatting stage. The proposed mechanism was applied on structurally complex data from four different domains. In order to test the validity of the proposal, we have applied data mining techniques to extract knowledge from the prepared data. The sound results of applying our proposal to several different domains confirm that it is applicable to any XML data source, as well as being correct, computationally efficient and saving time during the data preparation phase. [Copyright &y& Elsevier]