Identifying noisy features with the Pairwise Attribute Noise Detection Algorithm

计算机辅助诊断系统中,使用医学数据库的数据挖掘和知识发现技术具有重要意义和其发展的必然性.文章分析和综合了此类系统研究开发的一般规律及其特殊性,并指出了其中一些需要解决的问题和今后的研究发展方向.

Introduction Harvey J. Miller and Jiawei Han Spatiotemporal Data Mining Paradigms and Methodologies John F. Roddick and Brian G. Lees Fundamentals of Spatial Data Warehousing for Geographic Knowledge Discovery Yvan Bedard and Jiawei Han Analysis of Spatial Data with Map Cubes: Highway Traffic Data Chang-Tien Lu, Arnold P. Boedihardjo, and Shashi Shekhar NEW! Data Quality Issues and Geographic Knowledge Discovery Marc Gervais, Yvan Bedard, Marie-Andree Levesque, Eveline Bernier, and Rodolphe Devillers Spatial Classification and Prediction Models for Geospatial Data Mining Shashi Shekhar, Ranga Raju Vatsavai, and Sanjay Chawla An Overview of Clustering Methods in Geographic Data Analysis Jiawei Han, Jae-Gil Lee, and Micheline Kamber NEW! Computing Medoids in Large Spatial Datasets Kyriakos Mouratidis, Dimitris Papadias, Spiros Papadimitriou NEW! Looking for a Relationship? Try GWR A. Stewart Fotheringham, Martin Charlton, and Urska Demsar Leveraging the Power of Spatial Data Mining to Enhance the Applicability of GIS Technology Donato Malerba, Antonietta Lanza, and Annalisa Appice Visual Exploration and Explanation in Geography: Analysis with Light Mark Gahegan NEW! Multivariate Spatial Clustering and Geovisualization Diansheng Guo NEW! Toward Knowledge Discovery about Geographic Dynamics in Spatiotemporal Databases} May Yuan NEW! The Role of a Multitier Ontological Framework in Reasoning to Discover Meaningful Patterns of Sustainable Mobility Monica Wachowicz, Jose Macedo, Chiara Renso, and Arend Ligtenberg NEW! Periodic Pattern Discovery from Trajectories of Moving Objects Huiping Cao, Nikos Mamoulis, and David W. Cheung NEW! Decentralized Spatial Data Mining for Geosensor Networks Patrick Laube and Matt Duckham NEW! Beyond Exploratory Visualization of Space-Time Paths Menno-Jan Kraak and Otto Huisman

Formal Concept Analysis (FCA) is a valid tool for data mining and knowledge discovery, which identifies conceptual structures from (formal) contexts. As many practical applications involve non-bina...

The large number of government and industry activities supporting the Unit of Action (UA), with attendant documents, reports and briefings, can overwhelm decision-makers with an overabundance of information that hampers the ability to make quick decisions often resulting in a form of gridlock. In particular, the large and rapidly increasing amounts of data and data formats stored on UA Advanced Collaborative Environment (ACE) servers has led to the realization that it has become impractical and even impossible to perform manual analysis leading to timely decisions. UA Program Management (PM UA) has recognized the need to implement a Decision Support System (DSS) on UA ACE. The objective of this document is to research the commercial Knowledge Discovery and Data Mining (KDDM) market and publish the results in a survey. Furthermore, a ranking mechanism based on UA ACE-specific criteria has been developed and applied to a representative set of commercially available KDDM solutions. In addition, an overview of four R&D areas identified as critical to the implementation of DSS on ACE is provided. Finally, a comprehensive database containing detailed information on surveyed KDDM tools has been developed and is available upon customer request.

Telecommunications are one of the most dynamics and strategic areas in the world. In a complex and competitive scenario with pressure for better results with fewer resources, there is a constant need in the organizations for searching new and improved forms of administration with better understanding of the processes and customers. Techniques such as knowledge discovery in databases (KDD) and data mining (DM) appear as alternatives for enabling transform raw and complex data in models that may allow better decision making. This paper presents an application of multivariate regression analysis for modeling and explaining sales in the segment of corporate telecommunications, through acting indicators. A series of comparative analyses were accomplished with intend to identify the most appropriate or impacting indicators to perform the best decisions when selling corporate services. It is also presented a systemization of KDD activities which integrates the methodologies such as CRISPDM and SEMMA in an interactive design framework. Variable selection methods and statistical validation of the model are presented. Results are presented from real databases of large Brazilian telecommunications company.

The knowledge discovery and data mining (KDDM) process models describe the various phases (e.g. business understanding, data understanding, data preparation, modeling, evaluation and deployment) of the KDDM process. They act as a roadmap for implementation of the KDDM process by presenting a list of tasks for executing the various phases. The checklist approach of describing the tasks is not adequately supported by appropriate tools, which specify ‘how’ the particular task can be implemented. This may result in tasks not being implemented. Another disadvantage is that the long checklist does not capture or leverage the dependencies that exist among the various tasks of the same and different phases. This not only makes the process cumbersome to implement, but also hinders possibilities for semi-automation of certain tasks. Given that each task in the process model serves an important goal and even affects the execution of related tasks due to the dependencies, these limitations are likely to negatively affect the efficiency and effectiveness of KDDM projects. This paper proposes an improved KDDM process model that overcomes these shortcomings by prescribing tools for supporting each task as well as identifying and leveraging dependencies among tasks for semi-automation of tasks, wherever possible.

Analysing datasets using data mining techniques can enhance decision making in organizations. However, to ensure that the full potential of these techniques is realised it is important that decision makers understand there are Knowledge Discovery and Data Mining (KDDM) processes that are mature enough to be adopted. This paper demonstrates the benefits of using a KDDM process to evaluate survey data for internet banking users in Jamaica which includes demographic as well as attitudinal and behavioral variables. The major benefits of following this process include the selection of a set of models, rather than a single model, which are more relevant to the business/research objectives and use of a more targeted knowledge discovery process as the data mining analyst is now directed to consider the effects the decisions in each phase will have on subsequent phases. This leads to more relevant knowledge being extracted from the data mining process.

Preparing a data set for analysis is generally the most time consuming task in a data mining project, requiring many complex SQL queries, joining tables and aggregating columns. Existing SQL aggregations have limitations to prepare data sets because they return one column per aggregated group. In general, a significant manual effort is required to build data sets, where a horizontal layout is required. We propose simple, yet powerful, methods to generate SQL code to return aggregated columns in a horizontal tabular layout, returning a set of numbers instead of one number per row. This new class of functions is called horizontal aggregations. Horizontal aggregations build data sets with a horizontal denormalized layout (e.g. point-dimension, observation-variable, instance-feature), which is the standard layout required by most data mining algorithms. We propose three fundamental methods to evaluate horizontal aggregations: CASE: Exploiting the programming CASE construct; SPJ: Based on standard relational algebra operators (SPJ queries); PIVOT: Using the PIVOT operator, which is offered by some DBMSs. Experiments with large tables compare the proposed query evaluation methods. Our CASE method has similar speed to the PIVOT operator and it is much faster than the SPJ method. In general, the CASE and PIVOT methods exhibit linear scalability, where as the SPJ method does not. In a relational database, especially with normalized tables, a significant effort is required to prepare a summary data set that can be used as input for a data mining or statistical algorithm. Most algorithms require as input a data set with a horizontal layout, with several Records and one variable or dimension per column. That is the case with models like clustering, classification, regression and PCA; consult. Each research discipline uses different terminology to describe the data set. In data mining the common terms are point-dimension. Statistics literature generally uses observation-variable. Machine learning research uses instance-feature. This article introduces a new class of aggregate functions that can be used to build data sets in a horizontal layout (denormalized with aggregations), automating SQL query writing and extending SQL capabilities. We show evaluating horizontal aggregations is a challenging and interesting problem and we introduced alternative methods and optimizations for their efficient evaluation. II. MOTIVATION As mentioned above, building a suitable data set for data mining purposes is a time- consuming task. This task generally requires writing long SQL statements or customizing SQL Code if it is automatically generated by some tool. There are two main ingredients in such SQL code: joins and aggregations; we focus on the second one. The most widely- known aggregation is the sum of a column over groups of rows. Some other aggregations return the average, maximum, minimum or row count over groups of rows. There exist many aggregations functions and operators in SQL. Unfortunately, all these aggregations have limitations to build data sets for data mining purposes. The main reason is that, in general, data sets that are stored in a relational database (or a data warehouse) come

Data preparation is often cited as the most time consuming phase of a Knowledge Discovery and Data Mining (KDDM) process. This is attributed to the fact that this phase is highly dependent on the expertise of the analyst. Although process models exist for KDDM the description of their phases of the process focus on outlining what must be done but often do not detail how this should be done. While there is some research in addressing the how of the phases, the data preparation phase is thought to be the most challenging and is often described as an art rather than a science. The tasks defined in this phase are thought to be highly dependent on the expertise of the analyst and the context. While we are of the view that there will always be an art to data preparation we will demonstrate that the science can actually enhance the art. We further contend that as more research of this kind is published, that demonstrates a variety of data preparation techniques that enhance the data mining process, the more effective will be the science of data preparation.

New information and communication technologies are changing the way products are developed, manufactured, serviced and managed over the product’s lifecycle. Today’s smart products not only consist of their physical components, but are also endowed with intelligence. Data and the capabilities to process data into knowledge and eventually decisions have become critical components of the product itself and of the process to develop/operate the product. This paper investigates how engineers and a new functional role, data scientists, can effectively collaborate in a mixed team for new product development with data-driven features (NPD3). We focus on the concept development stage, typically the fuzziest phase of product development. In this paper, an integrated process model is explored by revisiting the traditional new product development (NPD) process model as well as the knowledge discovery and data mining (KDDM) process model. Then a case study of the development of an application-specific unmanned aircraft system (UAS) is used to examine the proposed model.

The 2003 ACM SIGMOD International Conference on Management of Data, was held June 10-12, 2003 in San Diego, California. This year, SIGMOD was part of the Federated Computing Research Conference (FCRC-2003), which included a total of 16 conferences and workshops. The SIGMOD conference is sponsored by the Association for Computing Machinery (ACM) and its Special Interest Group on Management of Data (SIGMOD). As has become customary, two days of the conference were overlapped with the Symposium on Principles of Database Systems (PODS). In addition, there were a number of important events co-located with SIGMOD this year, including the 6th International Workshop on the Web and Databases (WebDB), the Workshop on Research Issues in Data Mining and Knowledge Discovery (DMKD), The Second International Workshop on Distributed Event-Based Systems (MPDS), The Workshop on Management and Processing of Data Streams, the Paris C. Kanellakis Memorial Workshop, and the Second Database Faculty Symposium.As in previous years, SIGMOD held its status as a leading forum for exchange of ideas and publication of new results on data management. Acceptance into the conference proceedings was extremely competitive. We received 342 submissions to the conference (up from 240 in 2002, i.e., over 40% increase). The program committee selected 53 papers for presentation and inclusion in the proceedings. These papers span the range of traditional database topics such as query processing, indexing and OLAP, to current issues of signifcant interest such as XML data management, data streaming, large-scale data sharing, and metadata management. The program committee worked hard to select these papers through a detailed review process and active discussion both electronically and at the program committee meeting held in January in Seattle. We used the CMT to run the entire submissions and review process for research papers, including the program committee meeting.In addition to the research track, we had 73 submissions to the Demonstrations track, of which 22 projects were invited to present. The Demonstrations track has become a key venue for the early dissemination of cutting-edge prototype and systems development experience. Juliana Freire chaired the demos program committee, which put together a diverse and exciting program. The Industrial program committee, chaired by Surajit Chaudhuri, took an active role in soliciting quality submissions and have put together a program that is a key component of this year's conference: not only were there 4 papers accepted through the submission process, but there were an additional 10 invited talks. Likewise, Mary Fernandez and Joe Hellerstein organized an excellent slate of Tutorials and Panels to round out the program. As a result of all these efforts, we have an especially strong and innovative program this year, with a good balance of academic and industrial representation.

Nowadays, the development and influence of technology in human life is very important, where the role of technology greatly influences the activities carried out by humans. In a company organization, technology is not only used as a process to speed up the processes carried out. The use of such important technology also increases the size or volume of available data information. A dataset is a collection of data obtained in a data warehouse. Data mining is a technique that is part of Knowledge Discovery in Database (KDD). Clustering is a grouping process carried out in data mining. The first problem that is central to the research is that the values obtained from the clustering process are sometimes still not considered optimal. The performance results of the data mining clustering algorithm cannot yet be fully used as a basis for decision making. Comparisons made in clustering data mining are used to assist in the decision making process. In this research, the algorithms that will be used for comparison of performance are the K-Means and K-Medoids algorithms. Another problem that needs special attention is the problem of data quality. The results obtained from the data mining process can be seen from the quality of the data stored or used in the data processing process. Normalization is part of preprocessing data mining which aims to re-reason it based on a new scale. Z-Score is a normalization carried out on data based on statistical functions. The results obtained in the research The role of normalization in the research is very important, this is because using Z-Score normalization can improve the performance of the K-Means and K-Medoids algorithms, this can be seen from the DBI value obtained which is smaller when normalization is carried out compared to before it is carried out normalization, which indicates that performance is better after normalization. In the comparison of algorithms, the K-Medoids algorithm gets better performance, this can be seen from the DBI value obtained at 0.773 at K=9 after normalization. Meanwhile, the K-Means algorithm obtained a value of 0.783 at K=9 after normalization as well

As the number of patients treated in-home are increasing exponentially mainly in countries such as Japan, US A and Europe etc., these people often enter into a critical situation that may require help immediately (e.g. when facing an accident, or becoming depressed). Researches and Advances in computing areas and the Internet of Things (IOT) have provided efficient and cheap equipments including wireless communication and cameras, such as smart phones or embedded devices that enable the deployment of Health S mart Homes (HS H) that can provide medical treatment of patients in their homes. The images captured from these cameras can help nurses or caretakers of patients to provide timely help. The use of such patient images and emotional detection to assist patients and elderly people within home is provided in this article. Internet of Things and Cloud computing can work together, which can address to the Big Data problems. Big data is actually a term used for a huge data set that performs operations like storage of data, analysis, sharing, transfer, predictive analysis; updating etc. data set can grow rapidly. Data sets have different analytics on data that involve process of inspecting, transformation and modeling of data so as to discover new information, and to reach on some particular decision. There are some particular data analysis techniques from which data mining is a popular one which focuses on modeling and discovery of new facts. The discovered knowledge helps in predictive analysis purposes as well as text analysis. Further on data mining is performed on data set which involves discovery of patterns by applying different operations at databases, and using artificial intelligence, machine learning, statics etc. Keywords: Cloud computing, IOT, AI, face recognition, big data, data mining.

Five high-quality workshops were held at the 13th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2009) in Bangkok, Thailand during April 27-30, 2009. There were 17, 6, 9, 4 and 5 accepted papers to be presented at the Pacific Asia Workshop on Intelligence and Security Informatics (PAISI 2009), the workshop on Advances and Issues in Biomedical Data Mining (AIBDM 2009), the workshop on Data Mining with Imbalanced Classes and Error Cost (ICEC 2009), the workshop on Open Source in Data Mining (OSDM 2009), and the workshop on Quality Issues, Measures of Interestingness and Evaluation of Data Mining Models (QIMIE 2009). One competition, PAKDD 2009 Data Mining Competition, and one local workshop, Thai Track Session, were arranged. From these workshops (except PAISI which published its works in separate LNCS proceedings), we selected two or three best papers for this LNCS publication. PAKDD is a major international conference in the areas of data mining (DM) and knowledge discovery in database (KDD). It provides an international forum for researchers and industry practitioners to share their new ideas, original research results and practical development experiences from all KDD-related areas including data mining, data warehousing, machine learning, databases, statistics, knowledge acquisition and automatic scientific discovery,data visualization, causal induction and knowledge-based systems.

Most data mining algorithms and tools stop at discovered customer models, producing distribution information on customer profiles. Such techniques, when applied to industrial problems such as customer relationship management (CRM), are useful in pointing out customers who are likely attritors and customers who are loyal, but they require human experts to postprocess the discovered knowledge manually. Most of the postprocessing techniques have been limited to producing visualization results and interestingness ranking, but they do not directly suggest actions that would lead to an increase in the objective function such as profit. In this paper, we present novel algorithms that suggest actions to change customers from an undesired status (such as attritors) to a desired one (such as loyal) while maximizing an objective function: the expected net profit. These algorithms can discover cost-effective actions to transform customers from undesirable classes to desirable ones. The approach we take integrates data mining and decision making tightly by formulating the decision making problems directly on top of the data mining results in a postprocessing step. To improve the effectiveness of the approach, we also present an ensemble of decision trees which is shown to be more robust when the training data changes. Empirical tests are conducted on both a realistic insurance application domain and UCI benchmark data