Potential For Knowledge Discovery Research Articles

Improving River Routing Using a Differentiable Muskingum‐Cunge Model and Physics‐Informed Machine Learning

AbstractRecently, rainfall‐runoff simulations in small headwater basins have been improved by methodological advances such as deep neural networks (NNs) and hybrid physics‐NN models—particularly, a genre called differentiable modeling that intermingles NNs with physics to learn relationships between variables. However, hydrologic routing simulations, necessary for simulating floods in stem rivers downstream of large heterogeneous basins, had not yet benefited from these advances and it was unclear if the routing process could be improved via coupled NNs. We present a novel differentiable routing method (δMC‐Juniata‐hydroDL2) that mimics the classical Muskingum‐Cunge routing model over a river network but embeds an NN to infer parameterizations for Manning's roughness (n) and channel geometries from raw reach‐scale attributes like catchment areas and sinuosity. The NN was trained solely on downstream hydrographs. Synthetic experiments show that while the channel geometry parameter was unidentifiable, n can be identified with moderate precision. With real‐world data, the trained differentiable routing model produced more accurate long‐term routing results for both the training gage and untrained inner gages for larger subbasins (>2,000 km2) than either a machine learning model assuming homogeneity, or simply using the sum of runoff from subbasins. The n parameterization trained on short periods gave high performance in other periods, despite significant errors in runoff inputs. The learned n pattern was consistent with literature expectations, demonstrating the framework's potential for knowledge discovery, but the absolute values can vary depending on training periods. The trained n parameterization can be coupled with traditional models to improve national‐scale hydrologic flood simulations.

A systematic review on privacy-preserving distributed data mining

Combining and analysing sensitive data from multiple sources offers considerable potential for knowledge discovery. However, there are a number of issues that pose problems for such analyses, including technical barriers, privacy restrictions, security concerns, and trust issues. Privacy-preserving distributed data mining techniques (PPDDM) aim to overcome these challenges by extracting knowledge from partitioned data while minimizing the release of sensitive information. This paper reports the results and findings of a systematic review of PPDDM techniques from 231 scientific articles published in the past 20 years. We summarize the state of the art, compare the problems they address, and identify the outstanding challenges in the field. This review identifies the consequence of the lack of standard criteria to evaluate new PPDDM methods and proposes comprehensive evaluation criteria with 10 key factors. We discuss the ambiguous definitions of privacy and confusion between privacy and security in the field, and provide suggestions of how to make a clear and applicable privacy description for new PPDDM techniques. The findings from our review enhance the understanding of the challenges of applying theoretical PPDDM methods to real-life use cases, and the importance of involving legal-ethical and social experts in implementing PPDDM methods. This comprehensive review will serve as a helpful guide to past research and future opportunities in the area of PPDDM.

Dynamic resilience for biological wastewater treatment processes: Interpreting data for process management and the potential for knowledge discovery

Climate change, population growth and increasing regulation are causing wastewater treatment plants to become increasingly stressed, especially in countries like the UK, where many of these systems date back to the early part of the 20th century. Understanding resilience dynamics for these ageing wastewater assets represents a fundamental step in classifying multi-dimensional water stressors toward preventing severe pollution incidents. This paper explores the potential of a novel dynamic resilience approach to assess and predict the dynamic resilience of biological wastewater treatment based on the separation of stressor events (cause) and process stress (effect) to consider the deviation from reference conditions. The approach presented provides a fundamental link between (1) conventional activated sludge modelling methodologies, (2) actual biological wastewater process instrument data (potential for knowledge discovery) and (3) the characterisation of dynamic resilience in wastewater treatment processes. Results first present the dynamic resilience approach by modelling simulated shock flow conditions on an activated sludge plant, then incorporates ten years of wastewater process instrument data to demonstrate the actual dynamic resilience. The aim is to represent the “dynamic resilience” as self-ordering windows, a visual knowledge base (three dimensional, heat map), which operational staff can easily interpret. The outcomes presented suggest that such an approach is feasible and has the potential for real-time identification of conditions that result in pollution incidents based on actual historical process instrument data (knowledge discovery). Also, the methods presented could be extended to develop an improved understanding of wastewater system resilience under a range of future stressor scenarios.

Structured sparsity regularization for analyzing high-dimensional omics data.

The development of new molecular and cell technologies is having a significant impact on the quantity of data generated nowadays. The growth of omics databases is creating a considerable potential for knowledge discovery and, concomitantly, is bringing new challenges to statistical learning and computational biology for health applications. Indeed, the high dimensionality of these data may hamper the use of traditional regression methods and parameter estimation algorithms due to the intrinsic non-identifiability of the inherent optimization problem. Regularized optimization has been rising as a promising and useful strategy to solve these ill-posed problems by imposing additional constraints in the solution parameter space. In particular, the field of statistical learning with sparsity has been significantly contributing to building accurate models that also bring interpretability to biological observations and phenomena. Beyond the now-classic elastic net, one of the best-known methods that combine lasso with ridge penalizations, we briefly overview recent literature on structured regularizers and penalty functions that have been applied in biomedical data to build parsimonious models in a variety of underlying contexts, from survival to generalized linear models. These methods include functions of $\ell _k$-norms and network-based penalties that take into account the inherent relationships between the features. The successful application to omics data illustrates the potential of sparse structured regularization for identifying disease's molecular signatures and for creating high-performance clinical decision support systems towards more personalized healthcare. Supplementary information: Supplementary data are available at Briefings in Bioinformatics online.

Potential of knowledge discovery in automated test assembly

In the academic environment, acquired student's knowledge is usually assessed through paper- pencil or computer based tests. Testing is usually done several times during the academic year. Assembling a test that estimates students' knowledge requires significant examiner's effort and time invested. While assembling a test, examiner has to be objective, to provide consistent and reliable tests during the whole year. Automating the procedure can facilitate the process to the examiner and improve objectivity, consistency and reliability of student's knowledge assessment. This paper presents the results of development of a test assembly method based on the discovery of test assembly knowledge implicitly embedded in a bank of representative already administered tests. The method uses the nearest neighbour algorithm and it does not depend on a method of creation of existing tests in the test-bank. Results of the conducted experiments are promising and encourage further research in the same direction.

Facilitating design learning through faceted classification of in-service information

The maintenance and service records collected and maintained by engineering companies are a useful resource for the ongoing support of products. Such records are typically semi-structured and contain key information such as a description of the issue and the product affected. It is suggested that further value can be realised from the collection of these records for indicating recurrent and systemic issues which may not have been apparent previously. This paper presents a faceted classification approach to organise the information collection that might enhance retrieval and also facilitate learning from in-service experiences. The faceted classification may help to expedite responses to urgent in-service issues as well as to allow for patterns and trends in the records to be analysed, either automatically using suitable data mining algorithms or by manually browsing the classification tree. The paper describes the application of the approach to aerospace in-service records, where the potential for knowledge discovery is demonstrated.

Higher order mining

The value of knowledge obtainable by analysing large quantities of data is widely acknowledged. However, so-called primary or raw data may not always be available for knowledge discovery for several reasons. First, cooperating institutions that are interested in sharing knowledge may not be willing (or allowed) to disclose their primary data. Second, data in the form of streams are only temporarily available for processing. If stored at all, stream data are maintained in the form of synopses or derived, abstract representations of the original data. Finally, even for non-stream data, there are limits on the computation speed to be achieved -- such limits are set by hardware and firmware technologies. This problem can only be partially solved through parallelization and increased processing power. Ultimately, in many cases data must be summarized to be processed efficiently. In the light of these observations, we anticipate the need for defining and practising data mining without the luxury of primary data. To that end, we formally introduce the paradigm of Higher Order Mining as a form of data mining that is applied over non-primary, derived data or patterns. Although Higher Order Mining is a new paradigm, there are already research advances on knowledge discovery methods from patterns rather than data. We discuss them and organize them under the light of the new paradigm. We show that the HOM paradigm reveals further potential for knowledge discovery, including the delivery of rules and patterns with semantics that are closer to human intuition and are thus more appropriate for human inspection.

ESTIMATING DBLEARN'S POTENTIAL FOR KNOWLEDGE DISCOVERY IN DATABASES

We propose a procedure for estimating DBLEARN's potential for knowledge discovery, given a relational database and concept hierarchies. This procedure is most useful for evaluating alternative concept hierarchies for the same database. The DBLEARN knowledge discovery program uses an attribute‐oriented inductive‐inference method to discover potentially significant high‐level relationships in a database. A concept forest, with at most one concept hierarchy for each attribute, defines the possible generalizations that DBLEARN can make for a database. The potential for discovery in a database is estimated by examining the complexity of the corresponding concept forest. Two heuristic measures are defined based on the number, depth, and height of the interior nodes. Higher values for these measures indicate more complex concept forests and arguably more potential for discovery. Experimental results using a variety of concept forests and four commercial databases show that in practice both measures permit quite reliable decisions to be made; thus, the simplest may be most appropriate.