Mick J. Ridley

Data governance in predictive toxicology: A review.

BackgroundDue to recent advances in data storage and sharing for further data processing in predictive toxicology, there is an increasing need for flexible data representations, secure and consistent data curation and automated data quality checking. Toxicity prediction involves multidisciplinary data. There are hundreds of collections of chemical, biological and toxicological data that are widely dispersed, mostly in the open literature, professional research bodies and commercial companies. In order to better manage and make full use of such large amount of toxicity data, there is a trend to develop functionalities aiming towards data governance in predictive toxicology to formalise a set of processes to guarantee high data quality and better data management. In this paper, data quality mainly refers in a data storage sense (e.g. accuracy, completeness and integrity) and not in a toxicological sense (e.g. the quality of experimental results).ResultsThis paper reviews seven widely used predictive toxicology data sources and applications, with a particular focus on their data governance aspects, including: data accuracy, data completeness, data integrity, metadata and its management, data availability and data authorisation. This review reveals the current problems (e.g. lack of systematic and standard measures of data quality) and desirable needs (e.g. better management and further use of captured metadata and the development of flexible multi-level user access authorisation schemas) of predictive toxicology data sources development. The analytical results will help to address a significant gap in toxicology data quality assessment and lead to the development of novel frameworks for predictive toxicology data and model governance.ConclusionsWhile the discussed public data sources are well developed, there nevertheless remain some gaps in the development of a data governance framework to support predictive toxicology. In this paper, data governance is identified as the new challenge in predictive toxicology, and a good use of it may provide a promising framework for developing high quality and easy accessible toxicity data repositories. This paper also identifies important research directions that require further investigation in this area.

Promoting where, when and what? An analysis of web logs by integrating data mining and social network techniques to guide ecommerce business promotions

The rapid development of the internet introduced new trend of electronic transactions that is gradually dominating all aspects of our daily life. The amount of data maintained by websites to keep track of the visitors is growing exponentially. Benefitting from such data is the target of the study described in this paper. We investigate and explore the process of analyzing log data of website visitor traffic in order to assist the owner of a website in understanding the behavior of the website visitors. We developed an integrated approach that involves statistical analysis, association rules mining, and social network construction and analysis. First, we analyze the statistical data on the types of visitors that come to the website, as well as the steps they take to reach and satisfy the goal of their visit. Second, we derive association rules in order to identify the correlations between the web pages. Third, we study the links between the web pages by constructing a social network based on the frequency of access to the web pages such that two web pages get linked in the social network if they are identified as frequently accessed together. The value added from the overall analysis of the website and its related data should be considered valuable for ecommerce and commercial website owners; the owners will get the information needed to display targeted advertisements or messages to their customers. Such an automated approach gives advantage to its users in the current competitive cyberspace. In the long run, this is expected to allow for the increase in sales and overall customer loyalty.

Data warehouse architecture and design

A data warehouse is attractive as the main repository of an organization’s historical data and is optimized for reporting and analysis. In this paper, we present a data warehouse the process of data warehouse architecture development and design. We highlight the different aspects to be considered in building a data warehouse. These range from data store characteristics to data modeling and the principles to be considered for effective data warehouse architecture.

Predictive model representation and comparison: Towards data and predictive models governance

The increasing variety of data mining tools offers a large palette of types and representation formats for predictive models. Managing the models becomes then a big challenge, as well as reusing the models and keeping the consistency of model and data repositories because of the lack of an agreed representation across the models. The flexibility of XML representation makes it easier to provide solutions for Data and Model Governance (DMG) and support data and model exchange. We choose Predictive Toxicology as an application field to demonstrate our approach to represent predictive models linked to data for DMG. We propose an original structure: Predictive Toxicology Markup Language (PTML) offers a representation scheme for predictive toxicology data and models generated by data mining tools. We also show how this representation offers possibilities to compare models by similarity using our Distance Models Comparison technique. This work is ongoing and first encouraging results for calculating PTML distance are reported hereby.

Reporting and analyzing alternative clustering solutions by employing multi-objective genetic algorithm and conducting experiments on cancer data

Clustering is an essential research problem which has received considerable attention in the research community for decades. It is a challenge because there is no unique solution that fits all problems and satisfies all applications. We target to get the most appropriate clustering solution for a given application domain. In other words, clustering algorithms in general need prior specification of the number of clusters, and this is hard even for domain experts to estimate especially in a dynamic environment where the data changes and/or become available incrementally. In this paper, we described and analyze the effectiveness of a robust clustering algorithm which integrates multi-objective genetic algorithm into a framework capable of producing alternative clustering solutions; it is called Multi-objective K-Means Genetic Algorithm (MOKGA). We investigate its application for clustering a variety of datasets, including microarray gene expression data. The reported results are promising. Though we concentrate on gene expression and mostly cancer data, the proposed approach is general enough and works equally to cluster other datasets as demonstrated by the two datasets Iris and Ruspini. After running MOKGA, a pareto-optimal front is obtained, and gives the optimal number of clusters as a solution set. The achieved clustering results are then analyzed and validated under several cluster validity techniques proposed in the literature. As a result, the optimal clusters are ranked for each validity index. We apply majority voting to decide on the most appropriate set of validity indexes applicable to every tested dataset. The proposed clustering approach is tested by conducting experiments using seven well cited benchmark data sets. The obtained results are compared with those reported in the literature to demonstrate the applicability and effectiveness of the proposed approach.

Data retrieval from online social network profiles for social engineering applications

With the increased use of online social networking sites, data retrieval from social networking profiles is becoming a major tool for business. What makes social networking profile data different is its semi-structured format. The structure and the presentation of profile data change all the time. In social networking there is a lack of research into automated data retrieval from semi-structured web pages. Our approach is based on automated retrieval of the profiles attributes and list of top friends from MySpace by examining and extracting the relevant tokens in the parsed HTML code. The tokens were placed into a repository and Breadth First Search algorithm was used. The approach was implemented and tested with a profile which resulted in over 800 top friend profiles and attributes being extracted. This implementation process highlighted that MySpace profile structures vary depending on profile type and the way in which the user has customised the profile.

Reengineering XML into object-oriented database

This paper handles the conversion from an existing XML-Schema into object-oriented database. The major motivation for this work is to store XML-Schema into object oriented database. There are more common features between the object-oriented model and XML, and thus it is more attractive to map from XML into object-oriented database; such mapping preserves database specifics. To achieve the mapping, what we call the object graph is derived based on characteristics of the XML-Schema; it simply summarizes and includes all complex and simple elements and the links, which are the basics of the XML-Schema. Then, the links are simulated in terms of nesting to get a simulated object graph. This way, everything in a simulated object graph is directly representable in object-oriented database. Finally, we handle the mapping of the actual data from XML document(s) into the corresponding object-oriented database.

Algorithms for Data Retrieval from Online Social Network Graphs

In the last few years, data extraction from online social networks (OSNs) has become more automated. The aim of this study was to extract all friends from MySpace profiles in order to generate a friendship graph. The graph would be analysed to investigate and apply node vulnerability metrics. This research is an extension of our previous work which concentrated on the extraction of top friends but did not investigate the graph or node vulnerability. The graph was generated from the friendship links that were extracted and placed into a repository. From the graph structure and profiles’ personal details, vulnerability was calculated to find the most vulnerable node. Results were promising and provided interesting findings. Metric validation highlighted that the graph can be used to infer information that may not be present on the profile. The number of neighbours and the clustering coefficient were two main factors that affect the vulnerability of nodes.

Data governance strategy: a key issue in building Enterprise Data Warehouse

This paper articulates data governance as one of the key issue in building Enterprise Data Warehouse. The key goals of this document are to: define the strategy for Data Governance processes and procedures; define the scope of and identify major components of the data governance processes; adhere to enterprise Data Management standards, principles and guidelines; and articulate a vision for building, managing and safeguarding enterprise data foundation. The client-centric focus of business organizations coupled with aggressive attention to the bottom line propelled initiatives such as Data Governance to the top of the list of IT and business executives. The recent financial crisis which spawned the worldwide economic meltdown has been to a great extent blamed on non-trustworthy and non-transparent data. It is becoming progressively and patently evident that data MUST be managed like other assets such as financial and human resources. It has to have defined and mandated set of controls where compliance can be objectively measured and reported.

Transforming Object-Oriented Databases into XML

This paper presents a novel approach to transform an existing object-oriented database into XML. The major motivation to carry out this study is the fact that it is necessary to facilitate platform independent exchange of the content of object oriented databases. There are more common features between the object-oriented model and XML and thus the mapping from object-oriented databases into XML is less problematic. To achieve the mapping, what we call the object graph is derived based on characteristics of the object-oriented schema; it simply summarizes and includes all nesting and inheritance links, which are the basics of the object-oriented model. Then, the inheritance is simulated in terms of nesting to get a simulated object graph. This way, everything in a simulated object graph is directly representable in XML format. Finally, we handle the mapping of the actual data from the object-oriented database into corresponding XML document(s).