A combined neural network and DEA for measuring efficiency of large scale datasets

Data Envelopment Analysis (DEA) is one of the most widely used methods in the measurement of the efficiency and productivity of Decision Making Units (DMUs). DEA for a large dataset with many inputs/outputs would require huge computer resources in terms of memory and CPU time. This paper proposes a neural network back-propagation Data Envelopment Analysis to address this problem for the very large scale datasets now emerging in practice. Neural network requirements for computer memory and CPU time are far less than that needed by conventional DEA methods and can therefore be a useful tool in measuring the efficiency of large datasets. Finally, the back-propagation DEA algorithm is applied to five large datasets and compared with the results obtained by conventional DEA.

Parallel geostatistics for sparse and dense datasets

Very large spatially-referenced datasets, for example, those derived from satellite-based sensors which sample across the globe or large monitoring networks of individual sensors, are becoming increasingly common and more widely available for use in environmental decision making. In large or dense sensor networks, huge quantities of data can be collected over small time periods. In many applications the generation of maps, or predictions at specific locations, from the data in (near) real-time is crucial. Geostatistical operations such as interpolation are vital in this map-generation process and in emergency situations, the resulting predictions need to be available almost instantly, so that decision makers can make informed decisions and define risk and evacuation zones. It is also helpful when analysing data in less time critical applications, for example when interacting directly with the data for exploratory analysis, that the algorithms are responsive within a reasonable time frame. Performing geostatistical analysis on such large spatial datasets can present a number of problems, particularly in the case where maximum likelihood. Although the storage requirements only scale linearly with the number of observations in the dataset, the computational complexity in terms of memory and speed, scale quadratically and cubically respectively. Most modern commodity hardware has at least 2 processor cores if not more. Other mechanisms for allowing parallel computation such as Grid based systems are also becoming increasingly commonly available. However, currently there seems to be little interest in exploiting this extra processing power within the context of geostatistics. In this paper we review the existing parallel approaches for geostatistics. By recognising that diffeerent natural parallelisms exist and can be exploited depending on whether the dataset is sparsely or densely sampled with respect to the range of variation, we introduce two contrasting novel implementations of parallel algorithms based on approximating the data likelihood extending the methods of Vecchia [1988] and Tresp [2000]. Using parallel maximum likelihood variogram estimation and parallel prediction algorithms we show that computational time can be significantly reduced. We demonstrate this with both sparsely sampled data and densely sampled data on a variety of architectures ranging from the common dual core processor, found in many modern desktop computers, to large multi-node super computers. To highlight the strengths and weaknesses of the diffeerent methods we employ synthetic data sets and go on to show how the methods allow maximum likelihood based inference on the exhaustive Walker Lake data set.

Pragmatic algorithms for implementing geostatistics with large datasets

With the ability to collect and store increasingly large datasets on modern computers comes the need to be able to process the data in a way that can be useful to a Geostatistician or application scientist. Although the storage requirements only scale linearly with the number of observations in the dataset, the computational complexity in terms of memory and speed, scale quadratically and cubically respectively for likelihood-based Geostatistics. Various methods have been proposed and are extensively used in an attempt to overcome these complexity issues. This thesis introduces a number of principled techniques for treating large datasets with an emphasis on three main areas: reduced complexity covariance matrices, sparsity in the covariance matrix and parallel algorithms for distributed computation. These techniques are presented individually, but it is also shown how they can be combined to produce techniques for further improving computational efficiency.

GEO Label – quality information interrogation tool for geospatial datasets:towards effective visualization of quality metadata

Although the importance of dataset fitness-for-use evaluation and intercomparison is widely recognised within the GIS community, no practical tools have yet been developed to support such interrogation. GeoViQua aims to develop a GEO label which will visually summarise and allow interrogation of key informational aspects of geospatial datasets upon which users rely when selecting datasets for use. The proposed GEO label will be integrated in the Global Earth Observation System of Systems (GEOSS) and will be used as a value and trust indicator for datasets accessible through the GEO Portal. As envisioned, the GEO label will act as a decision support mechanism for dataset selection and thereby hopefully improve user recognition of the quality of datasets. To date we have conducted 3 user studies to (1) identify the informational aspects of geospatial datasets upon which users rely when assessing dataset quality and trustworthiness, (2) elicit initial user views on a GEO label and its potential role and (3), evaluate prototype label visualisations. Our first study revealed that, when evaluating quality of data, users consider 8 facets: dataset producer information; producer comments on dataset quality; dataset compliance with international standards; community advice; dataset ratings; links to dataset citations; expert value judgements; and quantitative quality information. Our second study confirmed the relevance of these facets in terms of the community-perceived function that a GEO label should fulfil: users and producers of geospatial data supported the concept of a GEO label that provides a drill-down interrogation facility covering all 8 informational aspects. Consequently, we developed three prototype label visualisations and evaluated their comparative effectiveness and user preference via a third user study to arrive at a final graphical GEO label representation. When integrated in the GEOSS, an individual GEO label will be provided for each dataset in the GEOSS clearinghouse (or other data portals and clearinghouses) based on its available quality information. Producer and feedback metadata documents are being used to dynamically assess information availability and generate the GEO labels. The producer metadata document can either be a standard ISO compliant metadata record supplied with the dataset, or an extended version of a GeoViQua-derived metadata record, and is used to assess the availability of a producer profile, producer comments, compliance with standards, citations and quantitative quality information. GeoViQua is also currently developing a feedback server to collect and encode (as metadata records) user and producer feedback on datasets; these metadata records will be used to assess the availability of user comments, ratings, expert reviews and user-supplied citations for a dataset. The GEO label will provide drill-down functionality which will allow a user to navigate to a GEO label page offering detailed quality information for its associated dataset. At this stage, we are developing the GEO label service that will be used to provide GEO labels on demand based on supplied metadata records. In this presentation, we will provide a comprehensive overview of the GEO label development process, with specific emphasis on the GEO label implementation and integration into the GEOSS.

Visualisation of bioinformatics datasets

Analysing the molecular polymorphism and interactions of DNA, RNA and proteins is of fundamental importance in biology. Predicting functions of polymorphic molecules is important in order to design more effective medicines. Analysing major histocompatibility complex (MHC) polymorphism is important for mate choice, epitope-based vaccine design and transplantation rejection etc. Most of the existing exploratory approaches cannot analyse these datasets because of the large number of molecules with a high number of descriptors per molecule. This thesis develops novel methods for data projection in order to explore high dimensional biological dataset by visualising them in a low-dimensional space. With increasing dimensionality, some existing data visualisation methods such as generative topographic mapping (GTM) become computationally intractable. We propose variants of these methods, where we use log-transformations at certain steps of expectation maximisation (EM) based parameter learning process, to make them tractable for high-dimensional datasets. We demonstrate these proposed variants both for synthetic and electrostatic potential dataset of MHC class-I. We also propose to extend a latent trait model (LTM), suitable for visualising high dimensional discrete data, to simultaneously estimate feature saliency as an integrated part of the parameter learning process of a visualisation model. This LTM variant not only gives better visualisation by modifying the project map based on feature relevance, but also helps users to assess the significance of each feature. Another problem which is not addressed much in the literature is the visualisation of mixed-type data. We propose to combine GTM and LTM in a principled way where appropriate noise models are used for each type of data in order to visualise mixed-type data in a single plot. We call this model a generalised GTM (GGTM). We also propose to extend GGTM model to estimate feature saliencies while training a visualisation model and this is called GGTM with feature saliency (GGTM-FS). We demonstrate effectiveness of these proposed models both for synthetic and real datasets. We evaluate visualisation quality using quality metrics such as distance distortion measure and rank based measures: trustworthiness, continuity, mean relative rank errors with respect to data space and latent space. In cases where the labels are known we also use quality metrics of KL divergence and nearest neighbour classifications error in order to determine the separation between classes. We demonstrate the efficacy of these proposed models both for synthetic and real biological datasets with a main focus on the MHC class-I dataset.

Rubric-Q:adding quality-related elements to the GEOSS clearinghouse datasets

Geospatial data have become a crucial input for the scientific community for understanding the environment and developing environmental management policies. The Global Earth Observation System of Systems (GEOSS) Clearinghouse is a catalogue and search engine that provides access to the Earth Observation metadata. However, metadata are often not easily understood by users, especially when presented in ISO XML encoding. Data quality included in the metadata is basic for users to select datasets suitable for them. This work aims to help users to understand the quality information held in metadata records and to provide the results to geospatial users in an understandable and comparable way. Thus, we have developed an enhanced tool (Rubric-Q) for visually assessing the metadata quality information and quantifying the degree of metadata population. Rubric-Q is an extension of a previous NOAA Rubric tool used as a metadata training and improvement instrument. The paper also presents a thorough assessment of the quality information by applying the Rubric-Q to all dataset metadata records available in the GEOSS Clearinghouse. The results reveal that just 8.7% of the datasets have some quality element described in the metadata, 63.4% have some lineage element documented, and merely 1.2% has some usage element described. © 2013 IEEE.

A knowledge driven approach towards the validation of externally acquired traceability datasets in supply chain business processes

The sharing of near real-time traceability knowledge in supply chains plays a central role in coordinating business operations and is a key driver for their success. However before traceability datasets received from external partners can be integrated with datasets generated internally within an organisation, they need to be validated against information recorded for the physical goods received as well as against bespoke rules defined to ensure uniformity, consistency and completeness within the supply chain. In this paper, we present a knowledge driven framework for the runtime validation of critical constraints on incoming traceability datasets encapuslated as EPCIS event-based linked pedigrees. Our constraints are defined using SPARQL queries and SPIN rules. We present a novel validation architecture based on the integration of Apache Storm framework for real time, distributed computation with popular Semantic Web/Linked data libraries and exemplify our methodology on an abstraction of the pharmaceutical supply chain.

Projected sequential Gaussian processes:a C++ tool for interpolation of large datasets with heterogeneous noise

Heterogeneous datasets arise naturally in most applications due to the use of a variety of sensors and measuring platforms. Such datasets can be heterogeneous in terms of the error characteristics and sensor models. Treating such data is most naturally accomplished using a Bayesian or model-based geostatistical approach; however, such methods generally scale rather badly with the size of dataset, and require computationally expensive Monte Carlo based inference. Recently within the machine learning and spatial statistics communities many papers have explored the potential of reduced rank representations of the covariance matrix, often referred to as projected or fixed rank approaches. In such methods the covariance function of the posterior process is represented by a reduced rank approximation which is chosen such that there is minimal information loss. In this paper a sequential Bayesian framework for inference in such projected processes is presented. The observations are considered one at a time which avoids the need for high dimensional integrals typically required in a Bayesian approach. A C++ library, gptk, which is part of the INTAMAP web service, is introduced which implements projected, sequential estimation and adds several novel features. In particular the library includes the ability to use a generic observation operator, or sensor model, to permit data fusion. It is also possible to cope with a range of observation error characteristics, including non-Gaussian observation errors. Inference for the covariance parameters is explored, including the impact of the projected process approximation on likelihood profiles. We illustrate the projected sequential method in application to synthetic and real datasets. Limitations and extensions are discussed. © 2010 Elsevier Ltd.