Principles for information visualization spreadsheets

The visualization spreadsheet provides a framework for exploring large and complex data sets. Structuring user interactions using a spreadsheet paradigm creates a powerful tool for information visualization     

Non-visual navigation of spreadsheets

The problem of non-visual navigation of the information in Excel? spreadsheet is that using current technologies, no overview is available for the user about different components found in the spreadsheet. Several attributes associated with spreadsheets make them not easy to navigate by individuals who are blind. The large amount of information stored in the spreadsheet, the multi-dimensional nature of the contents, and the several features it includes cannot be readily linearized by a screen reader or Braille display (e.g., charts and tables). A user-centered design paradigm is followed to build an accessible system for non-visual navigation of Microsoft Excel? spreadsheet. The proposed system provides the user with a hierarchical overview of the navigated components in an Excel? spreadsheet. The system is multi-modal, and it provides the user with different navigation and reading modes for the non-visual navigation of a spreadsheet. This will help the users in adapting the non-visual navigation according to the task the user needs to accomplish.     

Beyond spreadsheets: tools for building decision support systems

The complexity and long development time inherent in building decision support systems has thus far prevented their wide use. A new class of tools, DSS generators, seeks to cut the lead time between development and deployment. DSS generators provide tools that make it easier and faster to develop models, data, and user interfaces that are customized to the application's requirements. Using a DSS generator reduces DSS development to a decision analysis task-which requires expertise in decision analysis and mathematical modeling-rather than a programming task. DSS generators are crucial to the success of DSSs in practice. We describe the state of the art in DSS generator software, specifically in the realm of decision analysis methods. Decision analysis techniques account for the uncertain, dynamic, and multicriteria aspects of decisions. Essentially, they aid the evaluation of alternatives in the face of trade-offs. Well known decision analysis methods include decision trees and influence diagrams. We briefly describe the features of 11 commercially available DSS generators that specialize in decision analysis. Although not a comprehensive, complete analysis of these tools, the article clarifies the idea of DSS generators as DSS development environments and presents an overview of the progress in this area     

Canonical locality preserving Latent Variable Model for discriminative pose inference

Discriminative approaches for human pose estimation model the functional mapping, or conditional distribution, between image features and 3D poses. Learning such multi-modal models in high dimensional spaces, however, is challenging with limited training data; often resulting in over-fitting and poor generalization. To address these issues Latent Variable Models (LVMs) have been introduced. Shared LVMs learn a low dimensional representation of common causes that give rise to both the image features and the 3D pose. Discovering the shared manifold structure can, in itself, however, be challenging. In addition, shared LVM models are often non-parametric, requiring the model representation to be a function of the training set size. We present a parametric framework that addresses these shortcomings. In particular, we jointly learn latent spaces for both image features and 3D poses by maximizing the non-linear dependencies in the projected latent space, while preserving local structure in the original space; we then learn a multi-modal conditional density between these two low-dimensional spaces in the form of Gaussian Mixture Regression. With this model we can address the issue of over-fitting and generalization, since the data is denser in the learned latent space, as well as avoid the need for learning a shared manifold for the data. We quantitatively compare the performance of the proposed method to several state-of-the-art alternatives, and show that our method gives a competitive performance.     

Problem solving with spreadsheets*

The recent availability of spreadsheets has provided new opportunities for pupil and student use of microcomputers. Whilst much has been written on the use of database and word processing packages, the educational uses of spreadsheets is relatively undocumented. In this paper reference is made to exploratory work aimed at providing opportunities for additional experience in problem solving.     

Reasoning about spreadsheets with labels and dimensions

Labels in spreadsheets can be exploited for finding formula errors in two principally different ways. First, the spatial relationships between labels and other cells express simple constraints on the cells usage in formulas. Second, labels can be interpreted as units of measurements to provide semantic information about the data being combined in formulas, which results in different kinds of constraints.In this paper we demonstrate how both approaches can be combined into an integrated analysis, which is able to find significantly more errors in spreadsheets than each of the individual approaches. In particular, the integrated system is able to detect errors that cannot be found by either of the individual approaches alone, which shows that the integrated system provides an added value beyond the mere combination of its parts. We also compare the effectiveness of this combined approach with several other conceivable combinations of the involved components and identify a system that seems most effective to find spreadsheet formula errors based on label and unit-of-measurement information.     

Learning constraints in spreadsheets and tabular data

Spreadsheets, comma separated value files and other tabular data representations are in wide use today. However, writing, maintaining and identifying good formulas for tabular data and spreadsheets can be time-consuming and error-prone. We investigate the automatic learning of constraints (formulas and relations) in raw tabular data in an unsupervised way. We represent common spreadsheet formulas and relations through predicates and expressions whose arguments must satisfy the inherent properties of the constraint. The challenge is to automatically infer the set of constraints present in the data, without labeled examples or user feedback. We propose a two-stage generate and test method where the first stage uses constraint solving techniques to efficiently reduce the number of candidates, based on the predicate signatures. Our approach takes inspiration from inductive logic programming, constraint learning and constraint satisfaction. We show that we are able to accurately discover constraints in spreadsheets from various sources.     

Automatic detection of dimension errors in spreadsheets

We present a reasoning system for inferring dimension information in spreadsheets. This system can be used to check the consistency of spreadsheet formulas and thus is able to detect errors in spreadsheets.Our approach is based on three static analysis components. First, the spatial structure of the spreadsheet is analyzed to infer a labeling relationship among cells. Second, cells that are used as labels are lexically analyzed and mapped to potential dimensions. Finally, dimension information is propagated through spreadsheet formulas. An important aspect of the rule system defining dimension inference is that it works bi-directionally, that is, not only “downstream” from referenced arguments to the current cell, but also “upstream” in the reverse direction. This flexibility makes the system robust and turns out to be particularly useful in cases when the initial dimension information that can be inferred from headers is incomplete or ambiguous.We have implemented a prototype system as an add-in to Excel. In an evaluation of this implementation we were able to detect dimension errors in almost 50% of the investigated spreadsheets, which shows (i) that the system works reliably in practice and (ii) that dimension information can be well exploited to uncover errors in spreadsheets.     

Learning spreadsheets: human instruction vs. computer-based instruction

Abstract

The increasing need to instruct students in the use of personal computer software, especially electronic spreadsheets, is placing greater demands on the already full university curriculum. A potential help in meeting these demands is the readily available computer-based software tutorials. In order to explore the feasibility of computer-based instruction as an alternative to human instruction, this research compares two modes of instruction, computer-based and human. An experiment was conducted with groups of business student subjects. The research results indicate no difference in students‘ attitude towards computer-based instruction and human instruction of spreadsheets. Students’ short-term recall of the software syntax being taught is as strong with computer-based instruction as with human instruction. These results were not affected by the level of personal computer experience that students had prior to the experiment. However, the ability to comprehend and immediately apply the software to a task is greater with human instruction than with computer-aided instruction. This advantage holds true for students instructed by experienced and inexperienced instructors.     

Self-instructive spreadsheets: an environment for automatic knowledge acquisition and tutor generation

Typically, spreadsheet applications are difficult to use for casual users (different from developers), mainly because of lack of support. In fact, building a tutoring facility for such applications is a time-consuming task. Our aim is the realization of a tool for the automatic generation of Intelligent Tutors for conventional spreadsheet applications. We have developed a system that works in two steps. In the first one, it extracts an explicit representation of the problem-solving pattern coded in a programmed spreadsheet. In the second step, it generates a hypertextual guide and an interactive tutor that can effectively support, in the native environment, the casual user of the spreadsheet with the specific application it is designed for. We have successfully tested our system on a class of students using an application for budget analysis.     

Graphical inference for Infovis

How do we know if what we see is really there? When visualizing data, how do we avoid falling into the trap of apophenia where we see patterns in random noise? Traditionally, infovis has been concerned with discovering new relationships, and statistics with preventing spurious relationships from being reported. We pull these opposing poles closer with two new techniques for rigorous statistical inference of visual discoveries. The "Rorschach" helps the analyst calibrate their understanding of uncertainty and "line-up" provides a protocol for assessing the significance of visual discoveries, protecting against the discovery of spurious structure.     

Teaching statistics with Excel 2007 and other spreadsheets

This article considers which activities in teaching statistics may be suitable candidates for the application of spreadsheets, and whether spreadsheets in general and Excel 2007 in particular are suitable for these tasks.     

Bayesian inference for differential equations

Nonlinear dynamic systems such as biochemical pathways can be represented in abstract form using a number of modelling formalisms. In particular differential equations provide a highly expressive mathematical framework with which to model dynamic systems, and a very natural way to model the dynamics of a biochemical pathway in a deterministic manner is through the use of nonlinear ordinary or time delay differential equations. However if, for example, we consider a biochemical pathway the constituent chemical species and hence the pathway structure are seldom fully characterised. In addition it is often impossible to obtain values of the rates of activation or decay which form the free parameters of the mathematical model. The system model in many cases is therefore not fully characterised either in terms of structure or the values which parameters take. This uncertainty must be accounted for in a systematic manner when the model is used in simulation or predictive mode to safeguard against reaching conclusions about system characteristics that are unwarranted, or in making predictions that are unjustifiably optimistic given the uncertainty about the model. The Bayesian inferential methodology provides a coherent framework with which to characterise and propagate uncertainty in such mechanistic models and this paper provides an introduction to Bayesian methodology as applied to system models represented as differential equations.     

Depth-based inference for functional data

Robust inference tools for functional data are proposed. They are based on the notion of depth for curves. The ideas of trimmed regions, contours and central regions are extended to functions and their structural properties and asymptotic behavior are studied. Next, a scale curve is introduced to describe dispersion in a sample of functions. The computational burden of these techniques is not heavy, so they are also adequate to analyze high-dimensional data. These inferential methods are applied to several real data sets.     

Knowledge-guided inference for voice-enabled CAD

Voice based human-computer interactions have raised much interest and found various applications. Some extant voice based interactions only support voice commands with fixed vocabularies or preset expressions. This paper is motivated to investigate an approach to implement a flexible voice-enabled CAD system, where users are no longer constrained by predefined commands. Designers can, to a much more flexible degree, communicate with CAD modelers using natural language conversations. To accomplish this, a knowledge-guided approach is proposed to infer the semantics of voice input. The semantic inference is formulated as a template matching problem, where the semantic units parsed from voice input are the “samples” to be inspected and the semantic units in the predefined library are the feature templates. The proposed behavioral glosses, together with CAD-specific synonyms, hyponyms and hypernyms are extensively used in the parsing of semantic units and the subsequent template matching. Using such sources of knowledge, all the semantically equivalent expressions can be mapped to the same command set, and the Voice-enabled Computer Aided Design (VeCAD) system is then capable of processing new expressions it has never encountered and inferring/understanding the semantics at runtime. Experiments show that this knowledge-guided approach is helpful to enhance the robustness of semantic inference and can effectively eliminate the chance of overestimations and underestimations in design intent interpretation.     

Change-based inference for invariant discrimination

Under a conventional view of information processing in recurrently connected populations of neurons, computations consist in mapping inputs onto terminal attractor states of the dynamical interactions. However, there is evidence that substantial information representation and processing can occur over the course of the initial evolution of the dynamical states of such populations, a possibility that has attractive computational properties. Here, we suggest a model that explores one such property, namely, the invariance to an irrelevant feature dimension that arises from monitoring not the state of the population, but rather (a statistic of) the change in this state over time. We illustrate our proposal in the context of the bisection task, a paradigmatic example of perceptual learning for which an attractor-state recurrent model has previously been suggested. We show a change-based inference scheme that achieves near optimal performance in the task (with invariance to translation), is robust to high levels of dynamical noise and variations of the synaptic weight matrix, and indeed admits a computationally straightforward learning rule.     

Toward inference attacks for k-anonymity

Current research still cannot effectively prevent an inference attacker from inferring privacy information for k-anonymous data sets. To solve the issue, we must first study all kinds of aggressive reasoning behaviors and process for the attacker thoroughly. Our work focuses on describing comprehensively the inference attack and analyzing their privacy disclosures for k-anonymous data sets. In this paper, we build up a privacy inference graph based on attack graph theory, which is an extension of attack graph. The privacy inference graph describes comprehensively the inference attack in k-anonymous databases by considering attacker background knowledge and external factors. In the privacy inference graph, we introduce a concept of valid inference path to analyze the privacy disclosures in face of inference attack. According to both above, we design an algorithm to compute the n-valid inference paths. These paths can deduce some privacy information resulting in privacy disclosure. Moreover, we study the optimal privacy strategies to resist inference attack by key attribute sets and valid inference paths in the attack graph. An approximate algorithm is designed to obtain the approximate optimal privacy strategy set. At last, we prove the correctness in theory and analyze the performance of the approximate algorithm and their time complexity.     

Learning spreadsheets: human instruction vs. computer-based instruction

The increasing need to instruct students in the use of personal computer software, especially electronic spreadsheets, is placing greater demands on the already full university curriculum. A potential help in meeting these demands is the readily available computer-based software tutorials. In order to explore the feasibility of computer-based instruction as an alternative to human instruction, this research compares two modes of instruction, computer-based and human. An experiment was conducted with groups of business student subjects. The research results indicate no difference in students' attitude towards computer-based instruction and human instruction of spreadsheets. Students' short-term recall of the software syntax being taught is as strong with computer-based instruction as with human instruction. These results were not affected by the level of personal computer experience that students had prior to the experiment. However, the ability to comprehend and immediately apply the software to a task is greater with human instruction than with computer-aided instruction. This advantage holds true for students instructed by experienced and inexperienced instructors.