Search

This paper provides a unified description of Widening, a framework for the use of parallel (or otherwise abundant) computational resources to improve model quality. We discuss different theoretical approaches to Widening with and without consideration of diversity. We then soften some of the underlying constraints so that Widening can be implemented in real world algorithms. We summarize earlier experimental results demonstrating the potential impact as well as promising implementation strategies before concluding with a survey of related work.

This open access book constitutes the proceedings of the 18th International Conference on Intelligent Data Analysis, IDA 2020, held in Konstanz, Germany, in April 2020. The 45 full papers presented in this volume were carefully reviewed and selected from 114 submissions. Advancing Intelligent Data Analysis requires novel, potentially game-changing ideas. IDA’s mission is to promote ideas over performance: a solid motivation can be as convincing as exhaustive empirical evaluation. published

Widening is a method where parallel resources are used to find better solutions from greedy algorithms instead of merely trying to find the same solutions more quickly. To date, every example of Widening has used some form of communication between the parallel workers to maintain their distances from one another in the model space. For the first time, we present a communication-free, widened extension to a standard machine learning algorithm. By using Locality Sensitive Hashing on the Bayesian networks’ Fiedler vectors, we demonstrate the ability to learn classifiers superior to those of standard implementations and to those generated with a greedy heuristic alone.

New technologies to generate, store and retrieve medical and research data are inducing a rapid change in clinical and translational research and health care. Systems medicine is the interdisciplinary approach wherein physicians and clinical investigators team up with experts from biology, biostatistics, informatics, mathematics and computational modeling to develop methods to use new and stored data to the benefit of the patient. We here provide a critical assessment of the opportunities and challenges arising out of systems approaches in medicine and from this provide a definition of what systems medicine entails. Based on our analysis of current developments in medicine and healthcare and associated research needs, we emphasize the role of systems medicine as a multilevel and multidisciplinary methodological framework for informed data acquisition and interdisciplinary data analysis to extract previously inaccessible knowledge for the benefit of patients. Creating specialised computer systems to integrate big data from multiple sources may help provide tailored future healthcare. A healthcare system that meets the needs of every individual needs the ability to collect and mine data precisely and efficiently. Olaf Wolkenhauer at Rostock University, Germany and co-workers have explored the requirements for developing this ‘systems medicine’ approach. Interdisciplinary teams should be established specifically to develop software, computational models and integrated datasets for use in healthcare. Being able to harness information from huge medical datasets could dramatically improve our understanding of disease, and allow scientists to examine illnesses from the molecular scale to whole body responses. Systems medicine could also allow doctors to investigate an individual's specific needs, trial treatments through computer models, and analyze data captured on patients' smartphones, to provide safer, more personalised heathcare.

We demonstrate the application of Widening to learning performant Bayesian Networks for use as classifiers. Widening is a framework for utilizing parallel resources and diversity to find models in a hypothesis space that are potentially better than those of a standard greedy algorithm. This work demonstrates that widened learning of Bayesian Networks, using the Frobenius Norm of the networks’ graph Laplacian matrices as a distance measure, can create Bayesian networks that are better classifiers than those generated by popular Bayesian Network algorithms. published

Recently, the photomotor response (PMR) of zebrafish embryos was reported as a robust behavior that is useful for high-throughput neuroactive drug discovery and mechanism prediction. Given the complexity of the PMR, there is a need for rapid and easy analysis of the behavioral data. In this study, we developed an automated analysis workflow using the KNIME Analytics Platform and made it freely accessible. This workflow allows us to simultaneously calculate a behavioral fingerprint for all analyzed compounds and to further process the data. Furthermore, to further characterize the potential of PMR for mechanism prediction, we performed PMR analysis of 767 neuroactive compounds covering 14 different receptor classes using the KNIME workflow. We observed a true positive rate of 25% and a false negative rate of 75% in our screening conditions. Among the true positives, all receptor classes were represented, thereby confirming the utility of the PMR assay to identify a broad range of neuroactive molecules. By hierarchical clustering of the behavioral fingerprints, different phenotypical clusters were observed that suggest the utility of PMR for mechanism prediction for adrenergics, dopaminergics, serotonergics, metabotropic glutamatergics, opioids, and ion channel ligands. ispartof: Journal of Biomolecular Screening vol:21 issue:5 pages:427-436 ispartof: location:United States status: published

MS-based proteomics and metabolomics are rapidly evolving research fields driven by the development of novel instruments, experimental approaches, and analysis methods. Monolithic analysis tools perform well on single tasks but lack the flexibility to cope with the constantly changing requirements and experimental setups. Workflow systems, which combine small processing tools into complex analysis pipelines, allow custom-tailored and flexible data-processing workflows that can be published or shared with collaborators. In this article, we present the integration of established tools for computational MS from the open-source software framework OpenMS into the workflow engine Konstanz Information Miner (KNIME) for the analysis of large datasets and production of high-quality visualizations. We provide example workflows to demonstrate combined data processing and visualization for three diverse tasks in computational MS: isobaric mass tag based quantitation in complex experimental setups, label-free quantitation and identification of metabolites, and quality control for proteomics experiments. (VLID)2215121

In this paper we show that diversity-driven widening, the parallel exploration of the model space with focus on developing diverse models, can improve hierarchical agglomerative clustering. Depending on the selected linkage method, the model that is found through the widened search achieves a better silhouette coefficient than its sequentially built counterpart. published

We demonstrate that the previously introduced Widening framework is applicable to state-of-the-art Machine Learning algorithms. Using Krimp, an itemset mining algorithm, we show that parallelizing the search finds better solutions in nearly the same time as the original, sequential/greedy algorithm. We also introduce Reverse Standard Candidate Order (RSCO) as a candidate ordering heuristic for Krimp.

Given information about medical drugs and their properties, how can we automatically discover that Aspirin has blood-thinning properties, and thus prevents heart attacks? Expressed in more general terms, if we have a large in- formation network that integrates data from heterogeneous data sources, how can we extract semantic information that provides a better understanding of the integrated data and also helps us to identify missing links? We propose to extract concepts that describe groups of objects and their common properties from the integrated data. The discovered concepts provide semantic information as well as an abstract view on the integrated data and thus improve the understanding of complex systems. Our proposed method has the following desirable properties: (a) it is parameter-free and therefore requires no user-defined parameters (b) it is fault-tolerant, allowing for the detection of missing links and (c) it is scalable, being linear on the input size. We demonstrate the effectiveness and scalability of the proposed method on real, publicly available graphs.

This paper follows our earlier publicationi¾?[1], where we introduced the idea of tuned data mining which draws on parallel resources to improve model accuracy rather than the usual focus on speed-up. In this paper we present a more in-depth analysis of the concept of Widened Data Mining, which aims at reducing the impact of greedy heuristics by exploring more than just one suitable solution at each step. In particular we focus on how diversity considerations can substantially improve results. We again use the greedy algorithm for the set cover problem to demonstrate these effects in practice.

• COSMOS is at the heart of efforts to integrate different modelling approaches and chemoinformatics methods for in silico toxicity predictions to support the safety evaluation for cosmetic ingredients and beyond Introduction and Aims Andrea-Nicole Richarz1, Daniel Neagu2, Chihae Yang3, Elena Fioravanzo4, Alexandre RR Pery5, Andrew P Worth6, Michael R Berthold7, Mark TD Cronin1 Development of Integrated In Silico Models for Toxicity Prediction Focussing on Cosmetic Ingredients

We describe the application of a recently published general event detection framework, called EVE to the challenging task of molecular event detection, that is, the automatic detection of structural changes of a molecule over time. Different types of molecular events can be of interest which have, in the past, been addressed by specialized methods. The framework used here allows different types of molecular events to be systematically investigated. In this paper, we summarize existing molecular event detection methods and demonstrate how EVE can be configured for a number of molecular event types.

Representative members of the bioimage informatics community review the computational steps and some of the primary software tools available to biologists who are acquiring and analyzing microscopy-based digital image data, with a focus on open-source options. Few technologies are more widespread in modern biological laboratories than imaging. Recent advances in optical technologies and instrumentation are providing hitherto unimagined capabilities. Almost all these advances have required the development of software to enable the acquisition, management, analysis and visualization of the imaging data. We review each computational step that biologists encounter when dealing with digital images, the inherent challenges and the overall status of available software for bioimage informatics, focusing on open-source options.

We introduce a new, powerful query formulation formalism for complex, multivariate sequence data. The new query language, termed pattern graphs, is capable of reflecting more aspects of temporal patterns than earlier proposals. The underlying graph structure of the pattern graph makes the query intuitive to use and therefore understandable not only for the data analyst. We present algorithms to match patterns against data and demonstrate its usefulness on real data from the automobile industry.

In cellular systems environmental and metabolic signals are integrated for the conditional control of gene expression. On the other hand, artificial manipulation of gene expression is of high interest for metabolic and genetic engineering. Especially the reprogramming of gene expression patterns to orchestrate cellular responses in a predictable fashion is considered to be of great importance. Here we introduce a highly modular RNA-based system for performing Boolean logic computation at a post-transcriptional level in Escherichia coli. We have previously shown that artificial riboswitches can be constructed by utilizing ligand-dependent Hammerhead ribozymes (aptazymes). Employing RNA self-cleavage as the expression platform-mechanism of an artificial riboswitch has the advantage that it can be applied to control several classes of RNAs such as mRNAs, tRNAs, and rRNAs. Due to the highly modular and orthogonal nature of these switches it is possible to combine aptazyme regulation of activating a suppressor tRNA with the regulation of mRNA translation initiation. The different RNA classes can be controlled individually by using distinct aptamers for individual RNA switches. Boolean logic devices are assembled by combining such switches in order to act on the expression of a single mRNA. In order to demonstrate the high modularity, a series of two-input Boolean logic operators were constructed. For this purpose, we expanded our aptazyme toolbox with switches comprising novel behaviours with respect to the small molecule triggers thiamine pyrophosphate (TPP) and theophylline. Then, individual switches were combined to yield AND, NOR, and ANDNOT gates. This study demonstrates that post-transcriptional aptazyme-based switches represent versatile tools for engineering advanced genetic devices and circuits without the need for regulatory protein cofactors.

To enable discovery in large, heterogenious information networks a tool is needed that allows exploration in changing graph structures and integrates advanced graph mining methods in an interactive visualization framework. We present the Creative Exploration Toolkit (CET), which consists of a state-of-the-art user interface for graph visualization designed towards explorative tasks and support tools for integration and communication with external data sources and mining tools, especially the data-mining platform KNIME. All parts of the interface can be customized to fit the requirements of special tasks, including the use of node type dependent icons, highlighting of nodes and clusters. Through an evaluation we have shown the applicability of CET for structure-based analysis tasks.

We propose a two-phased approach to learn pattern graphs, a powerful pattern language for complex, multivariate temporal data, which is capable of reflecting more aspects of temporal patterns than earlier proposals. The first phase aims at increasing the understandability of the graph by finding common substructures, thereby helping the second phase to specialize the graph learned so far to discriminate against undesired situations. The usefulness is shown on data from the automobile industry and the libras data set by taking the accuracy and the knowledge gain of the learned graphs into account.

This article proposes a new approach to extract existing (or detect missing) concepts from a loosely integrated collection of information units by means of concept graph detection. Thereby a concept graph defines a concept by a quasi bipartite sub-graph of a bigger network with the members of the concept as the first vertex partition and their shared aspects as the second vertex partition. Once the concepts have been extracted they can be used to create higher level representations of the data. Concept graphs further allow the discovery of missing concepts, which could lead to new insights by connecting seemingly unrelated information units.

The discovery of surprising relations in large, heterogeneous information repositories is gaining increasing importance in real world data analysis. If these repositories come from diverse origins, forming different domains, domain bridging associations between otherwise weakly connected domains can provide insights into the data that are not accomplished by aggregative approaches. In this paper, we propose a first formalization for the detection of such potentially interesting, domain-crossing relations based purely on structural properties of a relational knowledge description.

Creative information exploration refers to a novel framework for exploring large volumes of heterogeneous information. In particular, creative information exploration seeks to discover new, surprising and valuable relationships in data that would not be revealed by conventional information retrieval, data mining and data analysis technologies. While our approach is inspired by work in the field of computational creativity, we are particularly interested in a model of creativity proposed by Arthur Koestler in the 1960s. Koestler's model of creativity rests on the concept of bisociation. Bisociative thinking occurs when a problem, idea, event or situation is perceived simultaneously in two or more "matrices of thought" or domains. When two matrices of thought interact with each other, the result is either their fusion in a novel intellectual synthesis or their confrontation in a new aesthetic experience. This article discusses some of the foundational issues of computational creativity and bisociation in the context of creative information exploration.

In this talk the author will discuss how parallel and/or distributed compute resources can be used differently: instead of focusing on speeding up algorithms, we propose to focus on improving accuracy. In a nutshell, the goal is to tune data mining algorithms to produce better results in the same time rather than producing similar results a lot faster. He will discuss a number of generic ways of tuning data mining algorithms and elaborate on two prominent examples in more detail. A series of examplatory experiments will be used to illustrate the effect such use of parallel resources can have.

Diversity selection is a common task in early drug discovery. One drawback of current approaches is that usually only the structural diversity is taken into account, therefore, activity information is ignored. In this article, we present a modified version of diversity selection, which we term Maximum-Score Diversity Selection, that additionally takes the estimated or predicted activities of the molecules into account. We show that finding an optimal solution to this problem is computationally very expensive (it is NP-hard), and therefore, heuristic approaches are needed. After a discussion of existing approaches, we present our new method, which is computationally far more efficient but at the same time produces comparable results. We conclude by validating these theoretical differences on several data sets.

This paper describes PMML extensions for the modular open source data analytics platform KNIME adding pre-processing support and the ability to edit existing PMML code. It is also shown how the PMML model representation in KNIME can be used within meta learning schemes such as boosting and bagging.

We present a supervised method for Learning in Parallel Universes, i.e. problems given in multiple descriptor spaces. The goal is the construction of local models in individual universes and their fusion to a superior global model that comprises all the available information from the given universes. We employ a predictive clustering approach using Neighborgrams, a one-dimensional data structure for the neighborhood of a single object in a universe. We also present an intuitive visualization, which allows for interactive model construction and visual comparison of cluster neighborhoods across universes.

In this paper we introduce a modular, highly flexible, open-source environment for data generation. Using an existing graphical data flow tool, the user can combine various types of modules for numeric and categorical data generators. Additional functionality is added via the data processing framework in which the generator modules are embedded. The resulting data flows can be used to document, deploy, and reuse the resulting data generators. We describe the overall environment and individual modules and demonstrate how they can be used for the generation of a sample, complex customer/product database with corresponding shopping basket data, including various artifacts and outliers.

In this paper we propose two methods to derive two different kinds of node similarities in a network based on their neighborhood. The first similarity measure focuses on the overlap of direct and indirect neighbors. The second similarity compares nodes based on the structure of their - possibly also very distant - neighborhoods. Instead of using standard node measures, both similarities are derived from spreading activation patterns over time. Whereas in the first method the activation patterns are directly compared, in the second method the relative change of activation over time is compared. We apply both methods to a real-world graph dataset and discuss the results.

We discuss Learning in parallel universes as a learning concept that encompasses the simultaneous analysis from multiple descriptor spaces. In contrast to existing approaches, this approach constructs a global model that is based on only partially applicable, local models in each descriptor space. We present some application scenarios and compare this learning strategy to other approaches on learning in multiple descriptor spaces. As a representative for learning in parallel universes we introduce different extensions to a family of unsupervised fuzzy clustering algorithms and evaluate their performance on an artificial data set and a benchmark of 3D objects.

We present a tool for interactive exploration of graphs that integrates advanced graph mining methods in an interactive visualization framework. The tool enables efficient exploration and analysis of complex graph structures. For flexible integration of state-of-the-art graph mining methods, the viewer makes use of the open source data mining platform KNIME. In contrast to existing graph visualization interfaces, all parts of the interface can be dynamically changed to specific visualization requirements, including the use of node type dependent icons, methods for a marking if nodes or edges and highlighting and a fluent graph that allows for iterative growing, shrinking and abstraction of (sub)graphs.

This paper is based on a panel discussion held at the Artificial Intelligence in Medicine Europe (AIME) conference in Amsterdam, The Netherlands, in July 2007. It had been more than 15 years since Edward Shortliffe gave a talk at AIME in which he characterized artificial intelligence (AI) in medicine as being in its "adolescence" (Shortliffe EH. The adolescence of Al in medicine: will the field come of age in the '90s? Artificial Intelligence in Medicine 1993;5:93-106). In this article, the discussants reflect on medical Al research during the subsequent years and characterize the maturity and influence that has been achieved to date. Participants focus on their personal areas of expertise, ranging from clinical decision-making, reasoning under uncertainty, and knowledge representation to systems integration, translational bioinformatics, and cognitive issues in both the modeling of expertise and the creation of acceptable systems. (C) 2008 Elsevier B.V. All rights reserved

The Konstanz Information Miner is a modular environment, which enables easy visual assembly and interactive execution of a data pipeline. It is designed as a teaching, research and collaboration platform, which enables simple integration of new algorithms and tools as well as data manipulation or visualization methods in the form of new modules or nodes. In this paper we describe some of the design aspects of the underlying architecture, briey sketch how new nodes can be incorporated, and highlight some of the new features of version 2.0.

In this paper we outline an approach for network-based information access and exploration. In contrast to existing methods, the presented framework allows for the integration of both semantically meaningful information as well as loosely coupled information fragments from heterogeneous information repositories. The resulting Bisociative Information Networks (BisoNets) together with explorative navigation methods facilitate the discovery of links across diverse domains. In addition to such "chains of evidence", they enable the user to go back to the original information repository and investigate the origin of each link, ultimately resulting in the discovery of previously unknown connections between information entities of different domains, subsequently triggering new insights and supporting creative discoveries.

Classifying large datasets without any a priori information poses a problem especially in the field of bioinformatics. In this work, we explore the problem of classifying hundreds of thousands of cell assay images obtained by a high-throughput screening camera. The goal is to label a few selected examples by hand and to automatically label the rest of the images afterwards. Up to now, such images are classified by scripts and classification techniques that are designed to tackle a specific problem. We propose a new adaptive active clustering scheme, based on an initial fuzzy c-means clustering and learning vector quantization. This scheme can initially cluster large datasets unsupervised and then allows for adjustment of the classification by the user. Motivated by the concept of active learning, the learner tries to query the most ''useful'' examples in the learning process and therefore keeps the costs for supervision at a low level. A framework for the classification of cell assay images based on this technique is introduced. We compare our approach to other related techniques in this field based on several datasets.

In this paper, we present a Dynamic Load Balancing (DLB) policy for problems characterized by a highly irregular search tree, whereby no reliable workload prediction is available. DLB approaches based on global statistics are known to provide optimal load balancing performance, while randomized techniques provide high scalability. The proposed method combines both advantages and adopts distributed job-pools and a randomized polling policy. The method has been successfully adopted in a parallel search algorithm for sugbraph mining. The work load distribution process of the parallel application is based on a dynamic partitioning of the search space and a peer-to-peer communication framework. The effectiveness of the DLB method has been evaluated on a molecular biology dataset. The distributed application with the novel DLB method has shown good scalability and close-to linear speedup in a distributed network of workstations. Moreover, fault tolerance and dynamic resource aggregation make it suitable for largescale, multi-domain, heterogeneous environments, such as computational Grids.

We present an extension of the fuzzy c-Means algorithm, which operates simultaneously on different feature spaces—so-called parallel universes—and also incorporates noise detection. The method assigns membership values of patterns to different universes, which are then adopted throughout the training. This leads to better clustering results since patterns not contributing to clustering in a universe are (completely or partially) ignored. The method also uses an auxiliary universe to capture patterns that do not contribute to any of the clusters in the real universes and therefore are likely to represent noise. The outcome of the algorithm is clusters distributed over different parallel universes, each modeling a particular, potentially overlapping subset of the data and a set of patterns detected as noise. One potential target application of the proposed method is biological data analysis where different descriptors for molecules are available but none of them by itself shows global satisfactory prediction results.

In this chapter we describe methods that assist the user to visually explore fuzzy clusters. We focus on a supervised approach to generate clusters for classes of interest of a given data set. The algorithm constructs local, one-dimensional neighborhood models, so-called Neighborgrams, for objects of the classes of interest that serve as a set of potential cluster candidates. The presented algorithm automatically chooses the best subset of Neighborgrams, but, more importantly, the accompanying visualization allows the user to fine-tune the clustering process by visually selecting, discarding, or adjusting potential cluster candidates. We also show how the algorithm can be applied to problems where multiple descriptions of data are available. This type of data can be found in biological data analysis for example, where often several different descriptors for the same molecule exist but each individual descriptor is only able to model parts of the data.

This paper presents two approaches to visually analyze the topic shift of a pool of documents over a given period of time. The first of the proposed methods is based on a multi-dimensional scaling algorithm, which places vectors representing terms occurring in certain years (period-frequency-vectors) in a spatial, two-dimensional space. This kind of visualization enables the detection of terms occurring in documents, published in particular years, or terms spread over different years. The second method uses a graph based approach. Publishing dates of documents, as well as their terms are represented by the vertices of a graph. Terms related to a specific publishing year are connected to the vertex of the year via an edge. By usage of activation spreading techniques, terms frequently occurring in documents published in particular years can be discovered visually. We tested both approaches with 2431 abstracts of papers published in the IEEE Transactions on SMC-A, SMC-B, and SMC-C in the years 1996 to 2006. Our experiments indicate that a number of interesting terms can be nicely separated in clumps according to individual years or periods of time. In addition, one can visualize the emergence of specific terms over certain periods of time and how these and other terms fade away again later.

Due to the nature of microarray experiments, gene expression levels across and through slide channels can experience up to 103 fold change differences in intensity. Such data variance is caused by `noise' elements, which can influence final expressions. This paper proposes a simple technique whereby histogram transformations are used to reduce noise artefacts. Akin to a magic eraser (removing the top layer of a surface), the technique attempts to blend pixels associated with gene spots into their background. The identification of pixels is relatively straightforward, but blending them with appropriate values is non-trivial. Once replacement values are determined, the background should be a good approximation of the original. By subtracting this surface from the original, gene spot regions would be more accurate. Experiments were carried out and results compared to “GenePix” a mainstream microarray process and “O'Neill” a microarray specific reconstruction algorithm. Not only was our process shown to be significantly quicker in execution time, it also reduced final expression results while typically generating less variation within gene's. This work is in part supported by EPSRC grant (EP/C524586/1).

The Konstanz Information Miner is a modular environment, which enables easy visual assembly and interactive execution of a data pipeline. It is designed as a teaching, research and collaboration platform, which enables simple integration of new algorithms and tools as well as data manipulation or visualization methods in the form of new modules or nodes. In this paper we describe some of the design aspects of the underlying architecture and briefly sketch how new nodes can be incorporated.

Analysis of simulation models has gained considerable interest in the past. However, their complexity still remains a considerable drawback in practical applications. A promising concept is to analyze the data from simulation experiments. Existing approaches are either restricted to simple models or are hard to interpret. We present an efficient algorithm that constructs a fuzzy graph model from simulation data and we show that the resulting system approximates also complex model functions with an adjustable precision. In addition the Fuzzy Graph allows the analyst to directly access easy to interpret if-then-rules. These rules help to understand the original simulation model, which is shown with a real world token bus model.