Data Mining Efforts Research Articles

Improvement Of Hubble Space Telescope Subsystems Through Data Mining

The Hubble Space Telescope continues to increase mankind’s knowledge and awareness of the universe. While in-orbit servicing has extended Hubble’s lifetime, the expense of servicing missions is extremely high. If mission lifetime is to be extended beyond the servicing mission era, innovative mission extension concepts must be employed. One such concept to accomplishing this is data mining of historical Hubble engineering data to increase the success rate of Hubble’s science data collection. A critical component of successfid science data collection is the Fixed Head Star Tracker (FHST) subsystem. This subsystem performs the fwst step, known as an update, in the telescope pointing process and has a failure rate of approximately 0.015. The NASA’s Hubble Project initiated a data mining effort on this subsystem, which was undertaken by the Institute for Scientific Research Inc. (ISR). Previous failure analyses indicated that many of the past failures could be attributed to the distribution of stars within the tracker’s field of view. A dataset was constructed of FHST data fi-om the following sources: Teleme@, Command, and Star data. This data was processed using a recursive-partitioning algorithm implemented in S-Plus. The resulting decision tree indicates that preevent FHST data supports failure prediction. Using this tree to select guide stars could reduce the tracker failure rate to 0.002. The decision tree has been validated using mission planning software. Data on the actual performance using this technique is being collected. This effort demonstrates how data mining can uncover operational deficiencies and provide cost-efficient solutions for improving subsystem performance. © 2002 WIT Press, Ashurst Lodge, Southampton, SO40 7AA, UK. All rights reserved. Web: www.witpress.com Email witpress@witpress.com Paper from: Data Mining III, A Zanasi, CA Brebbia, NFF Ebecken & P Melli (Editors). ISBN 1-85312-925-9

Confidentiality issues for medical data miners

The first task in any medical data mining effort is ensuring patient confidentiality. In the past, most data mining efforts ensured confidentiality by the dubious policy of witholding their raw data from colleagues and the public. A cursory review of medical informatics literature in the past decade reveals that much of what we have “learned” consists of assertions derived from confidential datasets unavailable for anyone’s review. Without access to the original data, it is impossible to validate or improve upon a researcher’s conclusions. Without access to research data, we are asked to accept findings as an act of faith, rather than as a scientific conclusion. This special issue of Artificial Intelligence in Medicine is devoted to medical data mining. The medical data miner has an obligation to conduct valid research in a way that protects human subjects. Today, data miners have the technical tools to merge large data collections and to distribute queries over disparate databases. In order to include patient-related data in shared databases, data miners will need methods to anonymize and deidentify data. This article reviews the human subject risks associated with medical data mining. This article also describes some of the innovative computational remedies that will permit researchers to conduct research AND share their data without risk to patient or institution.

Uniqueness of medical data mining

This article addresses the special features of data mining with medical data. Researchers in other fields may not be aware of the particular constraints and difficulties of the privacy-sensitive, heterogeneous, but voluminous data of medicine. Ethical and legal aspects of medical data mining are discussed, including data ownership, fear of lawsuits, expected benefits, and special administrative issues. The mathematical understanding of estimation and hypothesis formation in medical data may be fundamentally different than those from other data collection activities. Medicine is primarily directed at patient-care activity, and only secondarily as a research resource; almost the only justification for collecting medical data is to benefit the individual patient. Finally, medical data have a special status based upon their applicability to all people; their urgency (including life-or-death); and a moral obligation to be used for beneficial purposes.

Genomics and natural language processing.

The Human Genome and MEDLINE are both the foci of intense data-mining efforts worldwide. The biomedical literature has much to say about sequence, but it also seems that sequence can tell us much about the biomedical literature. Biological natural language processing is an emerging field of research that seeks to explore systematically the relationships between genes, sequences and the biomedical literature as a basis for a new generation of data-mining tools.

Benchmarking Data Mining Algorithms

Data mining is the process of sifting through the mass of organizational (internal and external) data to identify patterns critical for decision support. Successful implementation of the data mining effort requires a careful assessment of the various tools and algorithms available. The basic premise of this study is that machine-learning algorithms, which are assumption free, should outperform their traditional counterparts when mining business databases. The objective of this study is to test this proposition by investigating the performance of the algorithms for several scenarios. The scenarios are based on simulations designed to reflect the extent to which typical statistical assumptions are violated in the business domain. The results of the computational experiments support the proposition that machine learning algorithms generally outperform their statistical counterparts under certain conditions. These can be used as prescriptive guidelines for the applicability of data mining techniques.

Tissue Distribution and Functional Expression of a cDNA Encoding a Novel Mixed Lineage Kinase

Hypertrophy is an adaptive response of the heart to myocardial injury or hemodynamic overload that may progress and contribute to cardiac decompensation and eventually to heart failure. The signaling pathways controlling this response in the cardiac myocyte are poorly understood. A data mining effort of a human failed heart cDNA library was undertaken in an effort to identify novel signaling molecules involved in cardiac hypertrophy. This effort identified a novel kinase (MLK7) homologous to the mixed lineage kinase family of proteins. The mixed lineage kinases are mitogen-activated protein kinase kinase kinases (MAPKKKs) which activate stress activated protein kinase/c-Jun N-terminal kinase (SAPK/JNK) and p38 kinase pathways. They contain a catalytic domain with homology to both serine/threonine and tyrosine-specific kinases and a dual leucine zipper. MLK7 is identical to leucine zipper and sterile-alpha motif protein kinase (ZAK) through the leucine zipper domain but has a completely divergent COOH-terminus and shares approximately 40% homology with the other MLKs overall. Expression of MLK7 mRNA is most abundant in skeletal muscle and heart, with expression restricted to the cardiac myocyte. The recombinant histidine tagged MLK7 expressed and purified from insect cells exhibited serine/threonine kinase activity in vitro with myelin basic protein as substrate. When expressed in cardiac myocytes, MLK7 activated SAPK/JNK1, and ERK and p38 to a lesser extent. Additionally, MLK7 altered fetal gene expression and increased protein synthesis in cardiac myocytes. These data suggest that MLK7 is a new member of the mixed lineage kinase family that modulates cardiac SAPK/JNK pathway and may play a role in cardiac hypertrophy and progression to heart failure.

SPINE: an integrated tracking database and data mining approach for identifying feasible targets in high-throughput structural proteomics.

High-throughput structural proteomics is expected to generate considerable amounts of data on the progress of structure determination for many proteins. For each protein this includes information about cloning, expression, purification, biophysical characterization and structure determination via NMR spectroscopy or X-ray crystallography. It will be essential to develop specifications and ontologies for standardizing this information to make it amenable to retrospective analysis. To this end we created the SPINE database and analysis system for the Northeast Structural Genomics Consortium. SPINE, which is available at bioinfo.mbb.yale.edu/nesg or nesg.org, is specifically designed to enable distributed scientific collaboration via the Internet. It was designed not just as an information repository but as an active vehicle to standardize proteomics data in a form that would enable systematic data mining. The system features an intuitive user interface for interactive retrieval and modification of expression construct data, query forms designed to track global project progress and external links to many other resources. Currently the database contains experimental data on 985 constructs, of which 740 are drawn from Methanobacterium thermoautotrophicum, 123 from Saccharomyces cerevisiae, 93 from Caenorhabditis elegans and the remainder from other organisms. We developed a comprehensive set of data mining features for each protein, including several related to experimental progress (e.g. expression level, solubility and crystallization) and 42 based on the underlying protein sequence (e.g. amino acid composition, secondary structure and occurrence of low complexity regions). We demonstrate in detail the application of a particular machine learning approach, decision trees, to the tasks of predicting a protein's solubility and propensity to crystallize based on sequence features. We are able to extract a number of key rules from our trees, in particular that soluble proteins tend to have significantly more acidic residues and fewer hydrophobic stretches than insoluble ones. One of the characteristics of proteomics data sets, currently and in the foreseeable future, is their intermediate size ( approximately 500-5000 data points). This creates a number of issues in relation to error estimation. Initially we estimate the overall error in our trees based on standard cross-validation. However, this leaves out a significant fraction of the data in model construction and does not give error estimates on individual rules. Therefore, we present alternative methods to estimate the error in particular rules.

Data mining and serial documents

This paper is concerned with the investigation of the relevance and suitability of the data mining approach to serial documents. Conceptually the paper is divided into three parts. The first part presents the salient features of data mining and its symbiotic relationship to data warehousing. In the second part of the paper, historical serial documents are introduced, and the Ottoman Tax Registers (Defters) are taken as a case study. Their conformance to the data mining approach is established in terms of structure, analysis and results. A high-level conceptual model for the Defters is also presented. The final part concludes with a brief consideration of the implication of data mining for historical research.

Discovering Matrix Attachment Regions (MARs) in genomic databases

Lately, there has been considerable interest in applying Data Mining techniques to scientific and data analysis problems in bioinformatics. Data mining research is being fueled by novel application areas that are helping the development of newer applied algorithms in the field of bioinformatics, an emerging discipline representing the integration of biological and information sciences. This is a shift in paradigm from the earlier and the continuing data mining efforts in marketing research and support for business intelligence. The problem described in this paper is along a new dimension in DNA sequence analysis research and supplements the previously studied stochastic models for evolution and variability. The discovery of novel patterns from genetic databases as described is quite significant because biological pattern play an important role in a large variety of cellular processes and constitute the basis for gene therapy. Biological databases containing the genetic codes from a wide variety of organisms, including humans, have continued their exponential growth over the last decade. At the time of this writing, the GenBank database contains over 300 million sequences and over 2.5 billion characters of sequenced nucleotides. The focus of this paper is on developing a general data mining algorithm for discovering regions of locus control, i.e. those regions that are instrumental for activating genes. One type of such elements of locus control are the MARs or the Matrix Association Regions. Our limited knowledge about MARs has hampered their detection using classical pattern recognition techniques. Consequently, their detection is formulated by utilizing a statistical interestingness measure derived from a set of empirical features that are known to be associated with MARs. This paper presents a systematic approach for finding associations between such empirical features in genomic sequences, and for utilizing this knowledge in detecting biologically interesting control signals, such as MARs. This computational MAR discovery tool is implemented as a web-based software called MAR-Wiz and is available for public access. As our knowledge about the living system continues to evolve, and as the biological databases continue to grow, a pattern learning methodology similar to that described in this paper will be significant for the detection of regulatory signals embedded in genomic sequences.

When A Business Abandons A Data Mine: A Case Study

This case study examines the corporate expectations, conduct, and reactions surrounding a data-mining effort. Faced with the pressures of a still-restructuring business environment, even the executives who had initiated the effort failed to probe its value. They abandoned their commitment to prospecting and then the whole of the mining effort, despite its payback and promise. This article analyzes the project's failures and successes, offering guidelines that could make a difference in similar situations. If data mining is allowed to lose its strategic business focus, its potentialities will not be acknowledged.