Data representation

This paper proposes a method for the representation of functions describing a measured geophysical property (via sparsely scattered ordensely defined point data) by tensor and triple product interval B-splines (IBS). The spline representation facilitates archiving, data storage reduction, visualization and more general highlevel interrogation. Interval methods allow for the representation of the function values together with their uncertainty. The uncertainty is introduced, for example, because the measurement (or dependent variable) or the location of the sensor (the independent variable@)) is known only to a finite precision. In this paper, we present algorithms for the creation of IBS geometries based on minimization with linear constraints and we illustrate the method using geophysical ocean data and their interrogation. KEY WORDS: spline approximation; geophysical maps; interval methods 0 terrain modelingcreating a map from elevation data (or from contours) 0 CAD/CAM--creating computer models from measurements of the manufactured object; 0 computer vision or medical imagingcreating a visual surface or trivariate function from a variety of visual modalities (e.g. from photographs supplying shading, binocular stereo images, or tomography).

In the framework of indoor mobile robotics, place recognition is a challenging task, where it is crucial that self-localization be enforced precisely, notwithstanding the changing conditions of illumination, objects being shifted around and/or people affecting the appearance of the scene. In this scenario online learning seems the main way out, thanks to the possibility of adapting to changes in a smart and flexible way. Nevertheless, standard machine learning approaches usually suffer when confronted with massive amounts of data and when asked to work online. Online learning requires a high training and testing speed, all the more in place recognition, where a continuous flow of data comes from one or more cameras. In this paper we follow the Support Vector Machines-based approach of Pronobis et al. , proposing an improvement that we call Online Independent Support Vector Machines. This technique exploits linear independence in the image feature space to incrementally keep the size of the learning machine remarkably small while retaining the accuracy of a standard machine. Since the training and testing time crucially depend on the size of the machine, this solves the above stated problems. Our experimental results prove the effectiveness of the approach.

In this paper we present a new method, called TracMass, for analyzing data obtained using hyphenated chromatography-mass spectrometry (XC/MS). The method uses a Kalman filter to extract pure, noise-free ion chromatograms by exploiting the latent second order structure in the XC/MS data. TracMass differs from current state-of-the-art methodologies, which extract chromatograms by binning along the m/z axis and further processes the data in various ways, e.g. by baseline correction, component detection algorithm, peak detection, and curve resolution to extract molecular features. The proposed method was validated by analyzing two plasma datasets: one derived from 99 quality control samples where TracMass extracted 8880 Pure Ion Chromatograms (PICs) present in ≥90 of the samples. The second dataset was spiked with two different internal standard mixtures to test differential expression analysis. Here TracMass found 20 000 PICs present in 10 samples, all differentially expressed analytes, and also a previously unreported discriminating metabolite. Finding as many PICs as possible is in this context essential to ensure that even small differentiating features are found (if they exist). The resulting data representation from TracMass (PICs) can be used directly for statistical analysis, and the method is fast (approximately 5 min/sample), with few adjustable parameters.

The Free From Deformations are the most spread methods to modify the shape of geometrical objects, described with vertices and faces. Many applications have been developed to extend this approach, allowing different constraint specifications or actions. It has also been adapted to deform subdivision or implicit surfaces. This paper introduces an improvement for NURBS surfaces, allowing a point picked on the surface to follow a curvilinear path. The principle of the method is to let the user choose a point on the NURBS surface and its destination location in space. Then, the deformation range and the curvilinear path from the constraint point to the displacement goal must be specified. The proposed process computes the deformation of the initial object to match a displacement constraint. In order to be applied on NURBS surfaces, the process uses the (u, v) parameters of the initial chosen point, but this computational step is hidden for the user.

This paper presents an approach to product modelling that uses computational agents to represent product data. These agents are situated, i.e. they produce all data representations in response to the specific need in the current situation. We have explored this characteristic in the context of the communication of product data to develop a framework for agent-based product modelling. This framework complements current standardisation efforts through its ability to construct product models on the fly to flexibly adapt to changes in the environment.

This paper presents an approach to product modelling that uses computational agents to represent product data. These agents are situated, i.e. they produce all data representations in response to the specific need in the current situation. We have explored this characteristic in the context of the communication of product data to develop a framework for agent-based product modelling. This framework complements current standardisation efforts through its ability to construct product models on the fly to flexibly adapt to changes in the environment.

This paper presents an approach to product modelling that uses computational agents to represent product data. These agents are situated, i.e. they produce all data representations in response to the specific need in the current situation. We have explored this characteristic in the context of the communication of product data to develop a framework for agent-based product modelling. This framework complements current standardisation efforts through its ability to construct product models on the fly to flexibly adapt to changes in the environment.

development. Tony gave me several research topics to explore the research areas in which I am interested at the beginning, which finally led to my current thesis topic. After I decided on my research topic, he proposed many great ideas to improve my work. Tony would send me a couple of recently published papers related to my work almost every week, which broadened my mind very much. Tony sets very strict academic requirements for his students. But he is also a very generous friend in life. I enjoyed working with Tony closely over the past five years. This thesis would not have been possible without the help and support from my committee. They are Dr. Il-Yeol Song, Dr. Eileen Abels, Dr. Lisa Ulmer, Dr. Chaomei Chen, and Dr. Yuan An. I thank them for their great suggestions and comments on my thesis development. I really appreciate the time and work from Dr. Abels for her proof reading of my dissertation. I especially thank Dr. Song for his instructions and recommendations on my thesis organization, presentation skills and career development, which will benefit me in the future with no doubt. I am heartily thankful to Dr. Ulmer for her financial support. I really enjoyed working with her for her great leadership and personality.

This paper discusses a research study that focused on middle grades students' abilities to read and to move between different graphical representations before and after instruction. The data analyzed were collected from a group of 76 sixth grade students who were in 3 different mathematics classes in a middle school located in central North Carolina. The ways that students made sense of information presented through graphical representations and made connections between related pairs of graphs were investigated. General conclusions indicate that students need to talk more about graphs, including talking about the structure of graphs and how this informs the statements that can be made about: (1) the information depicted by the graphs and/or (2) the predictions or inferences that can be drawn from the graphs. Contains graphs and questions used on written pre- and posttests. Contains 26 references. (MKR)

Computational intelligence generally comprises a rather large set of -in a wider sense -adaptive and human-like data analysis and modelling methods. Due to some superior features -such as generalisation, trainability, coping with incomplete and inconsistent data, etc.computational intelligence has found its way into numerous applications in almost all scientific disciplines. A very prominent field among them are life sciences that are characterised by some unique requirements in terms of data structure and analysis.

In this paper, we give graphs whose topological index are exactly equal to the number $u_n$, satisfying the three term recurrence relation $$ u_n=a u_{n-1}+b u_{n-2}\quad(n\ge 2)\quad u_0=0\quad\hbox{and}\quad u_1=u\,, $$ where $a$, $b$ and $u$ are positive integers. We show an interpretation from the continued fraction expansion in a more general case, so that the topological index can be computed easily. On the contrary, for any given positive integer $N$, we can find the graphs (trees) whose topological indices are exactly equal to $N$.

FPGA technology has become widely used for real-time network intrusion detection. In this paper, a novel packet classification architecture called BV-TCAM is presented, which is implemented for an FPGA-based Network Intrusion Detection System (NIDS). The classifier can report multiple matches at gigabit per second network link rates. The BV-TCAM architecture combines the Ternary Content Addressable Memory (TCAM) and the Bit Vector (BV) algorithm to effectively compress the data representations and boost throughput. A tree-bitmap implementation of the BV algorithm is used for source and destination port lookup while a TCAM performs the lookup of the other header fields, which can be represented as a prefix or exact value. The architecture eliminates the requirement for prefix expansion of port ranges. With the aid of a small embedded TCAM, packet classification can be implemented in a relatively small part of the available logic of an FPGA. The design is prototyped and evaluated in a Xilinx FPGA XCV2000E on the FPX platform. Even with the most difficult set of rules and packet inputs, the circuit is fast enough to sustain OC48 traffic throughput. Using larger and faster FPGAs, the system can work at speeds greater than OC192.

At present, efforts are being made to pick up research data as bibliographic artifacts for re-use, transparency and citation. When approaching research data management solutions, it is imperative to consider carefully how filed data can be retrieved and accessed again on the user side. In the field of economics, a large amount of research is based on empirical data, which is often combined from several sources such as data centers, affiliated institutes or self-conducted surveys. Respecting this practice, we motivate and elaborate on techniques for fine-grained referencing of data fragments as to avoid multiple copies of same data archived over and over again, which may result in questionable transparency and difficult curation tasks. In addition, machines should have a deeper understanding of the given data, so that high-quality services can be installed. The paper first discusses the challenges of data management for the management of research data as used in empirical research. We conclude a comparison of referencing and copying strategies and reflect on their implications respectively. As a result from this argumentation, we elaborate on a data representation model, which we further examine in regard to considerable extensions. A Generating Model is subsequently introduced to enable citation, transparency and re-use. Eventually, we close with the demonstration of an * ACM, (2012). This is the authors version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definite version was published in i-KNOW '12 Proceedings of the 12th International Conference on Knowledge Management and Knowledge Technologies,

Query answers from on-line databases can easily be corrupted by hackers or malicious intent by the database publisher. Thus it is important to provide mechanisms which allow clients to trust the results from on-line queries. Authentic publication is a novel scheme which allows untrusted publishers to securely answer queries from clients on behalf of trusted o-line data owners. Publishers validate

Query answers from on-line databases can easily be corrupted by hackers or malicious intent by the database publisher. Thus it is important to provide mechanisms which allow clients to trust the results from on-line queries. Authentic publication is a novel scheme which allows untrusted publishers to securely answer queries from clients on behalf of trusted o-line data owners. Publishers validate

Biological motion has successfully been used for analysis of a person's mood and other psychological traits. Efforts are made to use human gait as a non-invasive mode of biometric. In this reported work, we try to study the effectiveness of biological gait motion of people as a cue to biometric based person recognition. The data is 3D in nature and, hence, has more information with itself than the cues obtained from videobased gait patterns. The high accuracies of person recognition, using a simple linear model of data representation and simple neighborhood based classfiers, suggest that it is the nature of the data which is more important than the recognition scheme employed.

Cognitive Radio Networks (CRNs) are emerging as a viable solution for exploiting the available spectrum and solving overcrowding in the operating bands. Several dimensions in the process of identification, management and routing need to be handled to overcome the challenges associated with the management of CRNs. A very important topic in achieving an effective and flexible management solution is related to the data representation of the different CRN dimensions. Scalability, high tolerance to churn and low latencies in accessing the stored data are fundamental characteristics in order to keep up with the dynamic nature of a mobile environment. Accordingly, to achieve a robust data representation that can handle all these requirements in an efficient way we focus our attention on peer-to-peer (P2P) systems. These systems are well known for providing an efficient platform to develop scalable and robust applications. Hence, in this paper we investigate using geometric multi-dimensional data representation and addressing system based on basic CAN functionality. Results obtained in our experiments are reported, providing useful information to improve the resilience to churn of our system.

Cognitive Radio Networks (CRNs) are expected to address and to resolve important technological and operational challenges of future networks, such as Dynamic Spectrum Access (DSA), routing in heterogeneous networks and provision of Quality of Experience (QoE) for different applications. Accordingly, a number of new terminal and network functionalities are required to solve the technical problems and to provide efficient management of CRNs. The paper advances a new solution for routing in CRNs, providing communication between Secondary Users (SUs) occupying different so-called Spectrum Opportunities (SOPs), i.e., portions of spectrum not occupied by Primary Users (PUs). This entails solving a complex process composed by two fundamental elements. These elements are multi-constraint routing and multi-dimensional adaptation for multiple cognitive radio dimensions like space, frequency, power and time. The goal of the paper is to develop optimization algorithms necessary to provide end-to-end (e2e) routing paths for communication between SUs with associated QoE demands.

The point of this study is to define the effect of project-based learning approach on 8th Grade secondary-school students' statistical literacy levels for data representation. To achieve this goal, a test which consists of 12 open-ended questions in accordance with the views of experts was developed. Seventy 8th grade secondary-school students, 35 in the experimental group and 35 in the control group, took this test twice, one before the application and one after the application. All the raw scores were turned into linear points by using the Winsteps 3.72 modelling program that makes the Rasch analysis and t-tests, and an ANCOVA analysis was carried out with the linear points. Depending on the findings, it was concluded that the project-based learning approach increases students' level of statistical literacy for data representation. Students' levels of statistical literacy before and after the application were shown through the obtained person-item maps.

This paper argues for a renewed focus on statistical reasoning in the elementary school years, with opportunities for children to engage in data modeling. Data modeling involves investigations of meaningful phenomena, deciding what is worthy of attention, and then progressing to organizing, structuring, visualizing, and representing data. Reported here are some findings from a two-part activity (Baxter Brown's Picnic and Planning a Picnic) implemented at the end of the second year of a current three-year longitudinal study (grade levels 1-3). Planning a Picnic was also implemented in a grade 7 class to provide an opportunity for the different age groups to share their products. Addressed here are the grade 2 children's predictions for missing data in Baxter Brown's Picnic, the questions posed and representations created by both grade levels in Planning a Picnic, and the metarepresentational competence displayed in the grade levels' sharing of their products for Planning a Picnic.

Identification of the navigational patterns of casual visitors is an important step in online recommendation to convert casual visitors to customers in e-commerce. Clustering and sequential analysis are two primary techniques for mining navigational patterns from Web and application server logs. The characteristics of the log data and mining tasks require new data representation methods and analysis algorithms to be tested in the e-commerce environment. In this paper we present a cube model to represent Web access sessions for data mining. The cube model organizes session data into three dimensions. The COMPO-NENT dimension represents a session as a set of ordered components {c1, c2, ..., cP }, in which each component ci indexes the ith visited page in the session. Each component is associated with a set of attributes describing the page indexed by it, such as the page ID, category and view time spent at the page. The attributes associated with each component are defined in the ATTRIBUTE dimension. The SESSION dimension indexes individual sessions. In the model, irregular sessions are converted to a regular data structure to which existing data mining algorithms can be applied while the order of the page sequences is maintained. A rich set of page attributes is embedded in the model for different analysis purposes. We also present some experimental results of using the partitional clustering algorithm to cluster sessions. Because the sessions are essentially sequences of categories, the k-modes algorithm designed for clustering categorical data and the clustering method using the Markov transition frequency (or probability) matrix, are used to cluster categorical sequences.

In this paper, we will start the discussion with the refinable generators of the shift invariant (SI) spaces in L 2 (R) that possess the largest possible regularities and required vanishing moments. For the pseudo-scaling generators, the corresponding MRA frame wavelets with certain regularities are constructed. In addition, the stability of the refinable SI spaces and the corresponding complementary spaces, biorthogonality of the SI spaces, and the approximation property of the spaces are also discussed.

Research on data integration has provided a set of rich and well understood schema mediation languages and systems that provide a meta-data representation of the modeled real world, while, in general, they do not deal with data instances. Such meta-data are necessary for querying classes result of an integration process: the end user typically does not know the contents of such classes, he simply defines his queries on the basis of the names of classes and attributes. In this paper we introduce an approach enriching the description of selected attributes specifying as meta-data a list of the "relevant values" for such attributes. Furthermore relevant values may be hierarchically collected in a taxonomy. In this way, the user may exploit new meta-data in the interactive process of creating/refining a query. The same meta-data are also exploited by the system in the query rewriting/unfolding process in order to filter the results showed to the user. We conducted an evaluation of the strategy in an e-business context within the EU-IST SEWASIE project. The evaluation proved the practicability of the approach for large value instances.

This paper presents the Adesso, a computational environment for the development of scientific software. The Adesso environment leverages the reusable software component programming model to support the development and integration of components to several scientific programming platforms. The Adesso system is based on an XML component database and a set of XML document transformation tools for the automatic generation of component code, documentation and packaging. An authoring tool, built with the help of the Adesso transformation system itself, is provided to assist the user in the creation of components. The Adesso system has been used for the development of several image processing toolboxes and applications. Some of them are presented briefly in this paper.

Data representation is a critical component in the field of data science and analytics, offering ways to convert raw data into a structured, interpretable format that can drive insights and support decision-making processes. This paper explores various techniques for data representation, including both traditional methods such as tables and charts, as well as modern, complex representations involving visualizations, graphs, and multidimensional data formats. The applications of these techniques span industries such as business, healthcare, and engineering, where insights derived from data representation influence strategies, enhance operational efficiencies, and predict future trends. By reviewing existing literature and implementing selected data representation techniques, this paper investigates both statistical and non-statistical methods to present data accurately and effectively. Through a detailed methodology and analysis, it further examines the effectiveness of these techniques in facilitating data-driven decision-making. This research ultimately provides recommendations for adopting optimal representation methods based on specific data characteristics and use cases.

The continuously increasing complexity of computer animations makes it necessary to rely on the knowledge of various experts to cover the different areas of computer graphics and animation. This fact, which can be noted in many areas of scientific working, leads to increasing effort being put into research concerning cooperative working over the internet. However, it still requires substantial effort and time to combine different animation techniques in a common virtual environment. When trying to perform collaborative animation over a network, we often face the problem of having to combine animation systems and applications based on different software and hardware and using incompatible data structures. We present an approach, based on a client-server architecture and employing a VRML-based language as common interchange format, that allows inhomogeneous systems to be easily incorporated into a collaborative animation. The applications can be freed from employing plug-ins or libraries to link into a common animation platform; they keep a local copy of the global scene and only need the ability to export the internal data representation into the so called "PaVRML" language, the language we use use to exchange data and synchronize the clients. This approach does not only allow a number of practitioners to share their know-how within a common animation without requiring the huge amount of work necessary to port their application to a common platform. It also makes it often possible in the first place to combine the capabilities of different animation systems into a single complex animation. Additionally, we investigate solutions to optimize the network load for real-time applications. In this paper we present preliminary results and discuss the future developments of this ongoing work.

The paper is organized as a self-contained literate Haskell program that implements elements of an executable finite set theory with focus on combinatorial generation and arithmetic encodings. The code, tested under GHC 6.6.1, is available at http://logic.csci.unt.edu/tarau/ research/2008/fSET.zip. We introduce ranking and unranking functions generalizing Ackermann's encoding to the universe of Hereditarily Finite Sets with Urelements. Then we build a lazy enumerator for Hereditarily Finite Sets with Urelements that matches the unranking function provided by the inverse of Ackermann's encoding and we describe functors between them resulting in arithmetic encodings for powersets, hypergraphs, ordinals and choice functions. After implementing a digraph representation of Hereditarily Finite Sets we define decoration functions that can recover well-founded sets from encodings of their associated acyclic digraphs. We conclude with an encoding of arbitrary digraphs and discuss a concept of duality induced by the set membership relation.

This paper is an exploration in a functional programming framework of isomorphisms between elementary data types (natural numbers, sets, multisets, finite functions, permutations binary decision diagrams, graphs, hypergraphs, parenthesis languages, dyadic rationals, primes, DNA sequences etc.) and their extension to hereditarily finite universes through hylomorphisms derived from ranking/unranking and pairing/unpairing operations. An embedded higher order combinator language provides any-to-any encodings automatically. Besides applications to experimental mathematics, a few examples of "free algorithms" obtained by transferring operations between data types are shown. Other applications range from stream iterators on combinatorial objects to self-delimiting codes, succinct data representations and generation of random instances. The paper covers 59 data types and, through the use of the embedded combinator language, provides 3540 distinct bijective transformations between them. The self-contained source code of the paper, as generated from a literate Haskell program, is available at http://logic.csci.unt.edu/tarau/ research/2008/fISO.zip. A short, 5 page version of the paper, published as [1] describes the idea of organizing various data transformations as encodings to sequences of natural numbers and gives a few examples of hylomorphisms that lift the encodings to related hereditarily finite universes.

We describe the BinProlog system's compilation technology, runtime system and its extensions supporting first-class Logic Engines while providing a short history of its development, details of some of its newer re-implementations as well as an overview of the most important architectural choices involved in their design. With focus on its differences with conventional WAM implementations, we explain key details of BinProlog's compilation technique, which replaces the WAM with a simplified continuation passing runtime system (the "BinWAM"), based on a mapping of full Prolog to binary logic programs. This is followed by a description of a term compression technique using a "tag-on-data" representation. Later derivatives, the Java-based Jinni Prolog compiler and the recently developed Lean Prolog system refine the BinProlog architecture with first-class Logic Engines, made generic through the use of an Interactor interface. An overview of their applications with focus on the ability to express at source level a wide variety of Prolog built-ins and extensions, covers these newer developments.

We address in this report the problem of representing formally textual data. First, this problem is replaced in the context of automatic text processing. Then, the weaknesses of the basic document representation, i.e. the bag-of-words representation, are explained and some state-ofthe-art methods claiming to overcome these weaknesses are reviewed. Moreover we propose a novel graphical model, the Theme Topic Mixture Model, which also claims to do so, in addition of giving a probabilistic framework in which documents are considered.

La industria agroalimentaria (IAA) debe de implantar un sistema de trazabilidad de los alimentos coordinado con los agentes que intervienen en el canal de distribución. En este trabajo se evalúan las relaciones IAA-Distribución en Canarias con objeto de conocer si presentan las condiciones adecuadas de

This paper is devoted to establishing sharp bounds for deviation probabilities of partial sums Σ n i=1 f(X i ), where X = (X n ) n✷◆ is a positive recurrent Markov chain and f is a real valued function defined on its state space. Combining the regenerative method to the Esscher transformation, these estimates are shown in particular to generalize probability inequalities proved in the i.i.d. case to the Markovian setting for (not necessarily uniformly) geometrically ergodic chains.

With the advent of wearable sensing and mobile technologies, biosignals have seen an increasingly growing number of application areas, leading to the collection of large volumes of data. One of the difficulties in dealing with these data sets, and in the development of automated machine learning systems which use them as input, is the lack of reliable ground truth information. In this paper we present a new web-based platform for visualization, retrieval and annotation of biosignals by non-technical users, aimed at improving the process of ground truth collection for biomedical applications. Moreover, a novel extendable and scalable data representation model and persistency framework is presented. The results of the experimental evaluation with possible users has further confirmed the potential of the presented framework.

In this paper we presented a violence detector built on the concept of visual codebooks using linear support vector machines. It differs from the existing works of violence detection in what concern the data representation, as none has considered local spatio-temporal features with bags of visual words. An evaluation of the importance of local spatio-temporal features for characterizing the multimedia content is conducted through the cross-validation method. The results obtained confirm that motion patterns are crucial to distinguish violence from regular activities in comparison with visual descriptors that rely solely on the space domain.

Computational methods have been used to study the effect of the molecular diffusivity of analytes on the band broadening and more particular the A-term band broadening, in liquid chromatography. The study was conducted on a two-dimensional mimic of a packed bed with nonporous particles (k = 0) and on a hypothetical perfectly ordered packed bed. The advantage of this computational approach is that the molecular diffusion coefficient of the solute in the mobile phase can be varied freely without affecting other parameters. The considered systems hence also formed a perfect test case for Giddings reduced plate height data representation method. The fact that the plate height values for the different D mol -cases were found to fall exactly on the same single reduced plate height curve hence simultaneously validates Giddings (undisputed) theory and the presently proposed method of chromatographic data simulation. The obtained results are also ideally suited for plate height model discrimination, because they yield highly accurate band broadening data (devoid of any experimental scatter) over a very broad range of reduced velocities (ν up to 9000). The best agreement between simulation data and theory is found for the Knox model in which the Knox coefficient was lowered from the widely used n = 1/3 to a value of n = 0.23. This value agrees well with recent data obtained by Popovici et al. in experimental plate height studies of size exclusion chromatography (SEC) of polymers in a range of reduced velocities amounting up to nearly ν = 10 5 .

The large amount of data collected today is quickly overwhelming researchers' abilities to interpret the data and discover interesting patterns. Knowledge discovery and data mining systems contain the potential to automate the interpretation process, but these approaches frequently utilize computationally expensive algorithms. In particular, scientific discovery systems focus on the utilization of richer data representation, sometimes without regard for scalability. This research investigates approaches for scaling a particular knowledge discovery data mining system, Subdue, using parallel and distributed resources. Subdue has been used to discover interesting and repetitive concepts in graph-based databases from a variety of domains, but requires a substantial amount of processing time. Experiments that demonstrate scalability of parallel versions of the Subdue system are performed using CAD circuit databases, satellite images, and artificially-generated databases, and potential achievements and obstacles are discussed. 2001

Feature-based modelling is the emerging technology to integrate design and engineering activities. It enriches product data representation with semantic information, that allows more efficient and direct communication between engineering processes. Nevertheless, major industrial problems arise from the fact that the communication between these processes is still not based on sharing the same information within a computer-internal product model. The lack of industrial product data exchange between various CAD/CAE applications is owing to the fact that there are diverse data formats and standards, giving rise to conversion problems. The ISO standard STEP provides a unique, comprehensive technology to solve most of the industrial problems in data exchange. However, it still does not provide all the resources needed to achieve a high-level, semantic data communication between engineering processes.In this article, we propose a solution to link the design process (CAD) with downstream engineering processes (CAE) based on an integrated feature-based modelling approach that supports both design-by-features and feature recognition. Integration is thereby achieved through a homogeneous, multiple view feature-based representation of the part model, called intermediate model, which is shared among the various applications.A distributed, object-oriented feature-based system architecture is described. Herein, the intermediate modeller server acts as a service provider to network-based distributed applications. These share the intermediate model and are able to extract their specific view from the semantic data of the product. A prototype implementation of the integrated feature-based modelling kernel and its integration into the distributed environment is also outlined.

Substantial evidence suggests that the phasic activities of dopamine neurons in the midbrain of primates report prediction errors in the delivery of reward. Recent recordings from these neurons in a task involving uncertain reward delivery present a crucial challenge to this hypothesis, and pose questions regarding the effects of both external and representational uncertainty on dopamine activity. Here, we analyse this issue, showing that the apparently anomalous activities are in fact what is expected under a standard prediction error account in light of different scalings of positive and negative prediction errors. We also study the implications of certain forms of representational noise on temporal difference learning.

Two-phase flow heat transfer has been exhaustively studied over recent years. However, in this field several questions remain unanswered. Heat transfer coefficient prediction related to nucleate and convective boiling have been studied using different approaches, numerical, analytical and experimental. In this work, an experimental analysis, data representation and heat transfer coefficient prediction on two-phase heat transfer on nucleate and convective boiling are presented. An empirical correlation is obtained based on genetic algorithms search engine over a dimensional analysis of the two-phase flow heat transfer problem.

Context: Recent advancements in the applicability of neural networks across a variety of fields, such as computer vision, natural language processing and others, have re-sparked an interest in program induction methods. (Kitzelman [1] , Gulwani et al. [2] or Kant [3].) Problem: When performing a program induction task, it is not feasible to search across all possible programs that map an input to an output because the number of possible combinations or sequences of instructions is too high: at least an exponential growth based on the generated program length. Moreover, there does not exist a general framework to formulate such program induction tasks and current computational limitations do not allow a very wide range of machine learning applications in the field of computer programs generation. Objective: In this study, we analyze the effectiveness of execution traces as learning representations for neural network models in a program induction setup. Our goal is to generate visualizations of program execution dynamics, specifically of sorting algorithms, and to apply machine learning techniques on them to capture their semantics and emulate their behavior using neural networks. Method: We begin by classifying images of execution traces for algorithms working on a finite array of numbers, such as various sorting and data structures algorithms. Next we experiment with detecting sub-program patterns inside the trace sequence of a larger program. The last step is to predict future steps in the execution of various sorting algorithms. More specifically, we try to emulate their behavior by observing their execution traces. We also discuss generalizations to other classes of programs, such as 1-D cellular automata. Results: Our experiments show that neural networks are capable of modeling the mechanisms underlying simple algorithms if enough execution traces are provided as data. We compare the performance of our program induction model with other similar experimental results from Graves et al. [4] and Vinyals et al. [5]. We were also able to demonstrate that sorting algorithms can be treated both as images displaying spatial patterns, as well as sequential instructions in a domain specific language, such as swapping two elements. We tested our approach on three types of increasingly harder tasks: detection, recognition and emulation. Conclusions: We demonstrate that simple algorithms can be modelled using neural networks and provide a method for representing specific classes of programs as either images or sequences of instructions in a domain-specific language, such that a neural network can learn their behavior. We consider the complexity of various setups to arrive at some improvements based on the data representation type. The insights from our experiments can be applied for designing better models of program induction.

Dynamic interactive technology brings new opportunities for helping students learn central statistical concepts. Research and classroom experience can be help identify concepts with which students struggle, and an "action-consequence" pre-made technology document can engage students in exploring these concepts. With the right questions, students can begin to make connections among their background in mathematics, foundational ideas that undergrid statistics and the relationship these ideas. The ultimate goal is to have students think deeply about simple and basic statistical ideas so they can see how they lead to reasoning and sense making about data and about making decisions about characteristics of a population from a sample.Technology has a critical role in teaching and learning statistics, enabling students to use real data in investigations, to model complex situations based on data, to visualize relationships using different representations, to move beyond calculations to interpreting statistical processes such as confidence intervals and correlation, and to generate simulations to investigate a variety of problems including laying a foundation for inference. Thus, graphing calculators, spreadsheets, and interactive dynamic software can all be thought of as tools for statistical sense making in the service of developing understanding. NEW OPPORTUNITIES Dynamic interactive technology has the potential to extend this tool to help students understand central statistical concepts. The ability to link representations, where changes in one representation are reflected in the others, enables students to take an action, immediately see the consequences and reflect on the meaning of these consequences to make sense of the statistics-an action-consequence principle (Dick & Burrill, 2006). To maximize this potential and allow students to explore statistical concepts in deeper ways, it is possible to impose constraints on what they can do, in essence creating action-consequence "microworlds" in which students can play with a statistical concept in a variety of ways but where the opportunity to go astray, both mathematically and operationally, is limited. An action-consequence document is similar to an applet (e.g

Deformable shape models (DSMs) comprise a general approach that shows great promise for automatic image segmentation. Published studies by others and our own research results strongly suggest that segmentation of a normal or near-normal object from 3D medical images will be most successful when the DSM approach uses 1) knowledge of the geometry of not only the target anatomic object but also the ensemble of objects providing context for the target object and 2) knowledge of the image intensities to be expected relative to the geometry of the target and contextual objects. The segmentation will be most efficient when the deformation operates at multiple object-related scales and uses deformations that include not just local translations but the biologically important transformations of bending and twisting, i.e., local rotation, and local magnification. In computer vision an important class of DSM methods uses explicit geometric models in a Bayesian statistical framework to provide a priori information used in posterior optimization to match the DSM against a target image. In this approach a DSM of the object to be segmented is placed in the target image data and undergoes a series of rigid and non-rigid transformations that deform the model to closely match the target object. The deformation process is driven by optimizing an objective function that has terms for the geometric typicality and model-to-image match for each instance of the deformed model. The success of this approach depends strongly on the object representation, i.e., the structural details and parameter set for the DSM, which in turn determines the analytic form of the objective function. This paper describes a form of DSM called mreps that has or allows these properties, and a method of segmentation consisting of large to small scale posterior optimization of m-reps. Segmentation by deformable m-reps, together with the appropriate data representations, visualizations, and user interface, has been implemented in software that accomplishes 3D segmentations in a few minutes. Software for building and training models has also been developed. The methods underlying this software and its abilities are the subject of this paper.

Deformable shape models (DSMs) comprise a general approach that shows great promise for automatic image segmentation. Published studies by others and our own research results strongly suggest that segmentation of a normal or near‐normal object from 3D medical images will be most successful when the DSM approach uses (1) knowledge of the geometry of not only the target anatomic object but also the ensemble of objects providing context for the target object and (2) knowledge of the image intensities to be expected relative to the geometry of the target and contextual objects. The segmentation will be most efficient when the deformation operates at multiple object‐related scales and uses deformations that include not just local translations but the biologically important transformations of bending and twisting, i.e., local rotation, and local magnification. In computer vision an important class of DSM methods uses explicit geometric models in a Bayesian statistical framework to provide...