PADMINI A Peer%2Dto%2DPeer Distributed Astronomy Data Mining System and a Case Study.

The next generation of transportation systems
Using immersive video to evaluate future
MineFleetr%3B%3A an oveview of  widely adopted distributed hicle performance data mining system.

A local asynchronous distributed privacy preserving feature selection algorithm for large peer%2Dto%2Dpeer networks.

ICDM 2009
Abstract—With the assistance of a domain expert, active learning can often select or construct fewer examples to request their labels to build an accurate classifier. However, previous works of active learning can only generate and ask specific queries. In real-world applications, the domain experts (or oracles) are often more readily to answer “generalized queries ” with don’t-care attributes. The power of such generalized queries is that one generalized query is often equivalent to many specific ones. However, overly general queries are not good as answers from the domain experts (or oracles) can be highly uncertain, and this makes learning difficult. In this paper, we propose a novel active learning algorithm that asks good generalized queries. We then extend our algorithm to construct new, hierarchical features for both nominal and numeric attributes. We demonstrate experimentally that, our new method asks significantly fewer queries compared with the previous works of active learning, even when the initial labeled dataset is very small, and the oracle is inaccurate in class probability estimations. Our method can be readily deployed in real-world data mining tasks where obtaining labeled examples is costly. This is an extension of the paper published in IEEE ICDM 2009 [16]. Keywords—active learning, domain expert, generalized query. 1
ICDM Workshops 2009

TagLearner A P2P Classifier Learning System from Collaboratively Tagged Text Documents.

Assured Information Sharing Life Cycle.

A Local Distributed Peer%2Dto%2DPeer Algorithm Using Multi%2DParty Optimization Based Privacy Preservation for Data Mining Primitive Computation.

Scalable Distributed Change Detection from Astronomy Data Streams Using Local
Electrical Engineering Surface curvature is used in a number of areas in computer graphics including mesh simplification, surface modeling and denoising, feature detection, and non-photorealistic line drawing techniques. Most applications must estimate surface curvature, due to the use of discrete models such as triangular meshes. Until recently, curvature estimation has been limited to CPU algorithms, forcing object geometry to reside in main memory. More computational work is being done directly on the GPU, and it is increasingly common for object geometry to
In%2DNetwork Outlier Detection in Wireless Sensor Networks
To address the problem of unsupervised outlier detection in wireless sensor networks, we develop an algorithm that (1) is flexible with respect to the outlier definition, (2) works in-network with a communication load proportional to the outcome, and (3) reveals its outcome to all sensors. We examine the algorithm’s performance using simulation with real sensor data streams. Our results demonstrate that the algorithm is accurate and imposes a reasonable communication load and level of power consumption. 1.
On the Privacy of Euclidean Distance Preserving Data Perturbation

Approximate Distributed K%2DMeans Clustering over a Peer%2Dto%2DPeer Network.

A Generic Local Algorithm for Mining Data Streams in Large Distributed Systems.
Abstract — In a large network of computers or wireless sensors, each of the components (henceforth, peers) has some data about the global state of the system. Much of the system’s functionality such as message routing, information retrieval and load sharing relies on modeling the global state. We refer to the outcome of the function (e.g., the load experienced by each peer) as the model of the system. Since the state of the system is constantly changing, it is necessary to keep the models up-to-date. Computing global data mining models e.g. decision trees, k-means clustering in large distributed systems may be very costly due to the scale of the system and due to communication cost, which may be high. The cost further increases in a dynamic scenario when the data changes rapidly. In this paper we describe a two step approach for dealing with these costs. First, we describe a highly efficient local algorithm which can be used to monitor a wide class of data mining models. Then, we use this algorithm as a feedback loop for the monitoring of complex functions of the data such as its k-means clustering. The theoretical claims are corroborated with a thorough experimental analysis. I.
A communication efficient probabilistic algorithm for mining frequent itemsets from a peer%2Dto%2Dpeer network.

Topic 5 Parallel and Distributed Databases.

Distributed Linear Programming and Resource Management for Data Mining in Distributed Environments.

Distributed Linear Programming and Resource Management for Data Mining in Distributed Environments.

An Efficient Local Algorithm for Distributed Multivariate Regression in Peer%2Dto%2DPeer Networks.
This paper offers a local distributed algorithm for multivariate regression in large peer-to-peer environments. The algorithm is designed for distributed inferencing, data compaction, data modeling and classification tasks in many emerging peer-to-peer applications for bioinformatics, astronomy, social networking, sensor networks and web mining. Computing a global regression model from data available at the different peer-nodes using a traditional centralized algorithm for regression can be very costly and impractical because of the large number of data sources, the asynchronous nature of the peer-to-peer networks, and dynamic nature of the data/network. This paper proposes a two-step approach to deal with this problem. First, it offers an efficient local distributed algorithm that monitors the “quality ” of the current regression model. If the model is outdated, it uses this algorithm as a feedback mechanism for rebuilding the model. The local nature of the monitoring algorithm guarantees low monitoring cost. Experimental results presented in this paper strongly support the theoretical claims. 1
Distributed Identification of Top%2Dl Inner Product Elements and its Application in a Peer%2Dto%2DPeer Network.
Abstract — Inner product measures how closely two feature vectors are related. It is an important primitive for many popular data mining tasks, e.g., clustering, classification, correlation computation, and decision tree construction. If the entire data set is available at a single site, then computing the inner product matrix and identifying the top (in terms of magnitude) entries is trivial. However, in many real-world scenarios, data is distributed across many locations and transmitting the data to a central server would be quite communication-intensive and not scalable. This paper presents an approximate local algorithm for identifying top-l inner products among pairs of feature vectors in a large asynchronous distributed environment such as a peer-to-peer (P2P) network. We develop a probabilistic algorithm for this purpose using order statistics and Hoeffding bound. We present experimental results to show the effectiveness and scalability of the algorithm. Finally, we demonstrate an application of this technique for interest-based community formation in a P2P environment. Index Terms — distributed data mining, inner product, peer-topeer network I.
Guest Editors%27 Introduction Special Section on Intelligence and Security Informatics.

Distributed probabilistic inferencing in sensor networks using variational approximation.

Distributed Decision%2DTree Induction in Peer%2Dto%2DPeer Systems.
This paper offers a scalable and robust distributed algorithm for decision tree induction in large Peer-to-Peer (P2P) environments. Computing a decision tree in such large distributed systems using standard centralized algorithms can be very communication-expensive and impractical because of the synchronization requirements. The problem becomes even more challenging in the distributed stream monitoring scenario where the decision tree needs to be updated in response to changes in the data distribution. This paper presents an alternate solution that works in a completely asynchronous manner in distributed environments and offers low communication overhead, a necessity for scalability. It also seamlessly handles changes in data and peer failures. The paper presents extensive experimental results to corroborate the theoretical claims. Index Terms peer-to-peer, data mining, decision trees I.
A Scalable Local Algorithm for Distributed Multivariate Regression.
This paper offers a local distributed algorithm for multivariate regression in large peer-to-peer environments. The algorithm can be used for distributed inferencing, data compaction, data modeling and classification tasks in many emerging peer-to-peer applications for bioinformatics, astronomy, social networking, sensor networks and web mining. Computing a global regression model from data available at the different peer-nodes using a traditional centralized algorithm for regression can be very costly and impractical because of the large number of data sources, the asynchronous nature of the peer-to-peer networks, and dynamic nature of the data/network. This paper proposes a two-step approach to deal with this problem. First, it offers an efficient local distributed algorithm that monitors the “quality ” of the current regression model. If the model is outdated, it uses this algorithm as a feedback mechanism for rebuilding the model. The local nature of the monitoring algorithm guarantees low monitoring cost. Experimental results presented in this paper strongly support the theoretical claims. Index Terms peer-to-peer, data mining, decision trees I.
A Scalable Local Algorithm for Distributed Multivariate Regression.
This paper offers a local distributed algorithm for multivariate regression in large peer-to-peer environments. The algorithm can be used for distributed inferencing, data compaction, data modeling and classification tasks in many emerging peer-to-peer applications for bioinformatics, astronomy, social networking, sensor networks and web mining. Computing a global regression model from data available at the different peer-nodes using a traditional centralized algorithm for regression can be very costly and impractical because of the large number of data sources, the asynchronous nature of the peer-to-peer networks, and dynamic nature of the data/network. This paper proposes a two-step approach to deal with this problem. First, it offers an efficient local distributed algorithm that monitors the “quality ” of the current regression model. If the model is outdated, it uses this algorithm as a feedback mechanism for rebuilding the model. The local nature of the monitoring algorithm guarantees low monitoring cost. Experimental results presented in this paper strongly support the theoretical claims. Index Terms peer-to-peer, data mining, decision trees I.
Uniform Data Sampling from a Peer%2Dto%2DPeer Network.
Uniform random sample is often useful in analyzing data. Usually taking a uniform sample is not a problem if the entire data resides in one location. However, if the data is distributed in a peer-to-peer (P2P) network with different amount of data in different peers, collecting a uniform sample of data becomes a challenging task. A random sampling can be performed using random-walk, but due to varying degrees of connectivity and different sizes of data owned by each peer, this random walk gives a biased sample. In this paper, we propose a random walk-based sampling algorithm that can be used to sample data tuples uniformly from a large, unstructured P2P network. We model the random walk as a Markov chain and derive conditions to bound the length of the random walk necessary to achieve uniformity. A formal communication analysis shows logarithmic communication cost to discover a uniform data sample. 1.
Multi%2Dparty
process communication have been examined from a complexity pbint of view [SP, Y]. We study a new model, in which a collection of processes eo, &amp;quot;&#039;&#039;, e~:~l
Distributed Top%2DK Outlier Detection from Astronomy Catalogs using the DEMAC System.

In%2DNetwork Outlier Detection in Wireless Sensor Networks.
To address the problem of unsupervised outlier detection in wireless sensor networks, we develop an algorithm that (1) is flexible with respect to the outlier definition, (2) works in-network with a communication load proportional to the outcome, and (3) reveals its outcome to all sensors. We examine the algorithm’s performance using simulation with real sensor data streams. Our results demonstrate that the algorithm is accurate and imposes a reasonable communication load and level of power consumption. 1.
An Attacker%27s View of Distance Preserving Maps for Privacy Preserving Data Mining.
Abstract. We examine the effectiveness of distance preserving transformations in privacy preserving data mining. These techniques are potentially very useful in that some important data mining algorithms can be efficiently applied to the transformed data and produce exactly the same results as if applied to the original data e.g. distance-based clustering, k-nearest neighbor classification. However, the issue of how well the original data is hidden has, to our knowledge, not been carefully studied. We take a step in this direction by assuming the role of an attacker armed with two types of prior information regarding the original data. We examine how well the attacker can recover the original data from the transformed data and prior information. Our results offer insight into the vulnerabilities of distance preserving transformations. 1
K%2DMeans Clustering Over a Large%2C Dynamic Network.

Local L2%2DThresholding Based Data Mining in Peer%2Dto%2DPeer Systems.
In a large network of computers, wireless sensors, or mobile devices, each of the components (hence, peers) has some data about the global status of the system. Many of the functions of the system, such as routing decisions, search strategies, data cleansing, and the assignment of mutual trust, depend on the global status. Therefore, it is essential that the system be able to detect, and react to, changes in its global status. Computing global predicates in such systems is usually very costly. Mainly because of their scale, and in some cases (e.g., sensor networks) also because of the high cost of communication. The cost further increases when the data changes rapidly (due to state changes, node failure, etc.) and computation has to follow these changes. In this paper we describe a two step approach for dealing with these costs. First, we describe a highly efficient local algorithm which detect when the L2 norm of the average data surpasses a threshold. Then, we use this algorithm as a feedback loop for the monitoring of complex predicates on the data – such as the data’s k-means clustering. The efficiency of the L2 algorithm guarantees that so long as the clustering results represent the data (i.e., the data is stationary) few resources are required. When the data undergoes an epoch change – a change in the underlying distribution – and the model no longer represents it, the feedback loop indicates this and the model is rebuilt. Furthermore, the existence of a feedback loop allows using approximate and “best-effort ” methods for constructing the model; if an ill-fit model is built the feedback loop would indicate so, and the model would be rebuilt. 1

Distributed Data Mining in Peer%2Dto%2DPeer Networks.
Distributed data mining deals with the problem of data analysis in environments with distributed data, computing nodes, and users. Peer-to-peer (P2P) computing is emerging as a new distributed computing paradigm for many novel applications that involve exchange of information among a large number of peers with little centralized coordination. P2P file sharing, P2P electronic commerce, and P2P monitoring based on a network of sensors are some examples. This paper offers an overview of distributed data mining applications and algorithms for P2P environments. It describes both exact and approximate distributed data mining algorithms that work in a decentralized manner.

Orthogonal Decision Trees.
This paper introduces orthogonal decision trees that offer an effective way to construct a redundancy-free, accurate, and meaningful representation of large decision-treeensembles often created by popular techniques such as Bagging, Boosting, Random Forests and many distributed and data stream mining algorithms. Orthogonal decision trees are functionally orthogonal to each other and they correspond to the principal components of the underlying function space. This paper offers a technique to construct such trees based on eigen-analysis of the ensemble and offers experimental results to document the performance of orthogonal trees on grounds of accuracy and model complexity. 1.
Clustering distributed data streams in peer%2Dto%2Dpeer environments.

On%2Dboard Vehicle Data Stream Monitoring Using MineFleet and Fast Resource Constrained Monitoring of Correlation Matrices.

Client%2Dside web mining for community formation in peer%2Dto%2Dpeer environments.
In this paper we present a framework for self-formation of interests-based Peer-to-Peer communities using client-side web browsing history. We propose an order statistics-based algorithm to build communities with hierarchical structures. We also carefully consider the privacy concerns of the peer and adopt cryptographic protocols to measure similarity between peers without disclosing their personal profiles. We have evaluated our algorithm using a distributed data mining toolkit. The experimental results show that our framework could efficiently build interests-based communities.
Topic 5 %2D Parallel and Distributed Databases

Orthogonal Decision Trees for Resource%2DConstrained Physiological Data Stream Monitoring Using Mobile Devices.
Several challenging new applications demand the ability to do data mining on resource constrained devices. One such application is that of monitoring physiological data streams obtained from wearable sensing devices. Such monitoring has applications for pervasive healthcare management, be it for seniors, emergency response personnel, soldiers in the battlefield or athletes. A key requirement is that the monitoring system be able to run on resource constrained handheld or wearable devices. Orthogonal decision trees (ODTs) offer an effective way to construct a redundancy-free, accurate, and meaningful representation of large decision-tree-ensembles often created by popular techniques such as Bagging, Boosting, Random Forests and many distributed and data stream mining algorithms. Orthogonal decision trees are functionally orthogonal to each other and they correspond to the principal components of the underlying function space. This paper discusses various properties of the ODTs and their suitability for monitoring physiological data streams in a resource-constrained environment. It offers experimental results to document the performance of orthogonal trees on grounds of accuracy, model complexity, and other characteristics in a resource-constrained mobile environment. 1.
Distributed data mining and agents.
Abstract. Multi-Agent Systems (MAS) offer an architecture for distributed problem solving. Distributed Data Mining (DDM) algorithms focus on one class of such distributed problem solving tasks—analysis and modeling of distributed data. This paper offers a perspective on DDM algorithms in the context of multiagents systems. It discusses broadly the connection between DDM and MAS. It provides a high-level survey of DDM, then focuses on distributed clustering algorithms and some potential applications in multi-agent-based problem solving scenarios. It reviews algorithms for distributed clustering, including privacypreserving ones. It describes challenges for clustering in sensor-network environments, potential shortcomings of the current algorithms, and future work accordingly. It also discusses confidentiality (privacy preservation) and presents a new algorithm for privacy-preserving density-based clustering.
Random%2Ddata perturbation techniques and privacy%2Dpreserving data mining.

Multi%2Dagent Systems and Distributed Data Mining.
We present a general Multi-Agent System framework for distributed data mining based on a Peer-to-Peer model. The framework adopts message-based asynchronous communication and a dynamic load balancing policy that is particularly suitable for irregular search algorithms. A modular design allows a separation of the general-purpose system protocols and software components from the specific data mining algorithm. While the general architecture has been implemented and successfully tested on a parallel frequent subgraph mining algorithm, several interesting issues still have to be explored. The present work will discuss them and will introduce the ongoing research efforts aimed at exploiting and leveraging the MAS for distributed data mining applications. 1 P2P-BASED MAS for Distributed Data Mining The last decade has seen an ever increasing availability of large amounts of data in many fields of science and in many IT applications. Data mining techniques have become popular techniques, which can reveal the valuable knowledge hidden in the rough data. However, these techniques often require high performance approaches in order to cope with the overwhelming amount of data and the complexity of algorithms. An effective approach is based on distributed and parallel computing.
Orthogonal Decision Trees.
This paper introduces orthogonal decision trees that offer an effective way to construct a redundancy-free, accurate, and meaningful representation of large decision-treeensembles often created by popular techniques such as Bagging, Boosting, Random Forests and many distributed and data stream mining algorithms. Orthogonal decision trees are functionally orthogonal to each other and they correspond to the principal components of the underlying function space. This paper offers a technique to construct such trees based on eigen-analysis of the ensemble and offers experimental results to document the performance of orthogonal trees on grounds of accuracy and model complexity. 1.
Privacy%2DSensitive Bayesian Network Parameter Learning.
This paper considers the problem of learning the parameters of a Bayesian Network, assuming the structure of the network is given, from a privacy-sensitive dataset that is distributed between multiple parties. For a binary-valued dataset, we show that the count information required to estimate the conditional probabilities in a Bayesian network can be obtained as a solution to a set of linear equations involving some inner product between the relevant different feature vectors. We consider a random projection-based method that was proposed elsewhere to securely compute the inner product (with a modified implementation of that method). 1.
Communication Efficient Construction of Decision Trees Over Heterogeneously Distributed Data.
We present an algorithm designed to efficiently construct a decision tree over heterogeneously distributed data without centralizing. We compare our algorithm against a standard centralized decision tree implementation in terms of accuracy as well as the communication complexity. Our experimental results show that by using only 20 % of the communication cost necessary to centralize the data we can achieve trees with accuracy at least 80 % of the trees produced by the centralized version.
VEDAS A Mobile and Distributed Data Stream Mining System for Real%2DTime Vehicle Monitoring.

A Fourier Spectrum%2DBased Approach to Represent Decision Trees for Mining Data Streams in Mobile Environments.
This paper presents a novel Fourier analysis-based technique to aggregate, transmit, and  visualize decision trees in a mobile environment. Fourier representation of a decision tree has  several interesting properties that are particularly useful for mining continuous data streams  from small mobile computing devices. This paper presents algorithms to compute the Fourier  spectrum of a decision tree and the vice versa. It offers a framework to aggregate decision  trees in their Fourier representations. It also describes MobiMine, a mobile data mining system  for mining stock-market data from handheld devices connected over low-bandwidth wireless  networks.
Learning Functions Using Randomized Genetic Code%2DLike Transformations Probabilistic Properties and Experimentations.

Collective Mining of Bayesian Networks from Distributed Heterogeneous Data.
We present a collective approach to learning a Bayesian network from distributed heterogenous data. In this approach, we first learn a local Bayesian network at each site using the local data. Then each site identifies the observations that are most likely to be evidence of coupling between local and non-local variables and transmits a subset of these observations to a central site. Another Bayesian network is learnt at the central site using the data transmitted from the local site. The local and central Bayesian networks are combined to obtain a collective Bayesian network, that models the entire data. Experimental results and theoretical justification that demonstrate the feasibility of our approach are presented.
On the Privacy Preserving Properties of Random Data Perturbation Techniques.
Privacy is becoming an increasingly important issue in many data mining applications. This has triggered the development of many privacy-preserving data mining techniques. A large fraction of them use randomized data distortion techniques to mask the data for preserving the privacy of sensitive data. This methodology attempts to hide the sensitive data by randomly modifying the data values often using additive noise. This paper questions the utility of the random value distortion technique in privacy preservation. The paper notes that random objects (particularly random matrices) have “predictable ” structures in the spectral domain and it develops a random matrix-based spectral filtering technique to retrieve original data from the dataset distorted by adding random values. The paper presents the theoretical foundation of this filtering method and extensive experimental results to demonstrate that in many cases random data distortion preserve very little data privacy. 1.
Towards a Pervasive Grid.

Privacy Sensitive Distributed Data Mining from Multi%2Dparty Data.

Analysis of privacy preserving random perturbation techniques further explorations.
Privacy is becoming an increasingly important issue in many data mining applications, particularly in the security and defense area. This has triggered the development of many privacy-preserving data mining techniques. A large fraction of them uses randomized data distortion techniques to mask the data for preserving the privacy. They attempt to hide the sensitive data by randomly modifying the data values using additive noise. This paper questions the utility of such randomized data distortion technique for preserving privacy in many cases and urges caution.It notes that random objects (particularly random matrices) have “predictable” structures in the spectral domain and then offers a random matrix-based spectral filtering technique to retrieve original data from the data-set distorted by adding random values. It extends our earlier work questioning the efficacy of random perturbation techniques using additive noise for privacy-preserving data mining in continuous valued domain and presents new results in the discrete domain. It shows that the growing collection of random perturbation-based “privacy-preserving ” data mining techniques may need a careful scrutiny in order to prevent privacy breaches through linear transformations. The paper also presents extensive experimental results in order to support this claim. This is an extension of the paper “Random Data Perturbation
Dependency detection in MobiMine a systems perspective.

A Random Matrix%2DBased Approach for Dependency Detection from Data Streams.

A Resampling Technique for Learning the Fourier Spectrum of Skewed Data.
Function induction using the widely studied Walsh or Multidimensional Discrete Fourier Transform (MDFT) coefficient estimates has several benefits, including the fact that decision trees can be constructed efficiently from the spectrum. While these estimates are accurate for uniform data, highly skewed data is the norm. This paper gives a way of improving the accuracy of the MDFT coefficient estimates in the case of skewed data. An adaptive resampling algorithm for learning the MDFT coefficients is presented and verified experimentally. An equivalent estimator that can be learned using a single pass algorithm is defined. The effectiveness of the technique is demonstrated using controlled experiments.
A Resampling Technique for Learning the Fourier Spectrum of Skewed Data.
Function induction using the widely studied Walsh or Multidimensional Discrete Fourier Transform (MDFT) coefficient estimates has several benefits, including the fact that decision trees can be constructed efficiently from the spectrum. While these estimates are accurate for uniform data, highly skewed data is the norm. This paper gives a way of improving the accuracy of the MDFT coefficient estimates in the case of skewed data. An adaptive resampling algorithm for learning the MDFT coefficients is presented and verified experimentally. An equivalent estimator that can be learned using a single pass algorithm is defined. The effectiveness of the technique is demonstrated using controlled experiments.
Constructing Simpler Decision Trees from Ensemble Models Using Fourier Analysis.
Ensemble learning is frequently used for classification and other related applications in data mining. It generates multiple models and produces the final classification by aggregating the outputs of the different models in the ensemble. However, large ensembles are often hard to interpret and difficult to translate into action-able knowledge. This paper considers the construction of a decision tree from the Fourier spectrum of an ensemble model within a user-defined range of errors. The Fourier spectrum of an ensemble of decision trees retains all the necessary information that can be used to construct a simpler &#034;informative&#034; decision tree. This approach can be effectively used for building ensemble-based classifiers from both static data sets and data streams.
Dependency Detection in MobiMine and Random Matrices.
This paper describes a novel approach to detect correlation  from data streams in the context of MobiMine --- an experimental mobile  data mining system. It presents a brief description of the MobiMine  and identifies the problem of detecting dependencies among stocks from  incrementally observed financial data streams. This is a non-trivial problem  since the stock-market data is inherently noisy and small incremental  volumes of data makes the estimation process more vulnerable to noise.
Distributed
This paper presents an algorithm by which a process in a distributed system determines a global state of the system during a computation. Many problems in distributed systems can be cast in terms of the problem of detecting global states. For instance, the global state detection algorithm helps to solve an important class of problems: stable property detection. A stable property is one that persists: once a stable property becomes true it remains true thereafter. Examples of stable properties are “computation has terminated, ”  “ the system is deadlocked ” and “all tokens in a token ring have disappeared. ” The stable property detection problem is that of devising algorithms to detect a given stable property. Global state detection can also be used for checkpointing. Categories and Subject Descriptors: C.2.4 [Computer-Communication Networks]: Distributed Systems-distributed applications; distributed databases; network operating systems; D.4.1 [Operating Systems]: Process Management-concurrency; deadlocks, multiprocessing/multiprogramming; mutual exclusion; scheduling; synchronization; D.4.5 [Operating Systems]: Reliability-backup procedures; checkpoint/restart; fault-tolerance; verification
Editorial Computation in Gene Expression.

Toward Machine Learning Through Genetic Code%2Dlike Transformations.

Book Reviews. Review of Advances in Distributed and Parallel Knowledge Discovery.

MobiMine Monitoring the Stock Market from a PDA.

Mining Decision Trees from Data Streams in a Mobile Environment.
This paper presents a novel Fourier analysis-based technique to aggregate, communicate, and visualize decision trees in a mobile environment. Fourier representation of a decision tree has several useful properties that are particularly useful for mining continuous data streams from small mobile computing devices. This paper presents algorithms to compute the Fourier spectrum of a decision tree and the vice versa. It offers a framework to aggregate decision trees in their Fourier representations. It also describes a touchpad /ticker-based approach to visualize decision trees using their Fourier spectrum and an implementation for PDAs.  1. 
Distributed Web Mining Using Bayesian Networks from Multiple Data Streams.


Data mining %22to go%22 ubiquitous KDD for mobile and distributed environments.
Spatial data mining is the discovery of interesting relationships and characteristics that may exist implicitly in spatial databases. In this paper, we explore whether clustering methods have a role to play in spatial data mining. To this end, we develop a new clustering method called CLARANS which is based on randomized search. We also de- velop two spatial data mining algorithms that use CLARANS. Our analysis and experiments show that with the assistance of CLARANS, these two algorithms are very effective and can lead to discoveries that are difficult to find with current spatial data mining algorithms.
Gene Expression and Fast Construction of Distributed Evolutionary Representation.
The gene expression process in nature produces different proteins in different cells from different portions of the DNA. Since proteins control almost every important activity in a living organism, at an abstract level, gene expression can be viewed as a process that evaluates the merit or &#034;fitness&#034; of the DNA. This distributed evaluation of the DNA would not be possible without a decomposed representation of the fitness function defined over the DNAs. This paper argues that, unless the living body was provided with such a representation, we have every reason to believe that it must have an efficient mechanism to construct this distributed representation. This paper demonstrates polynomial-time computability of such a representation by proposing a class of efficient algorithms. The main contribution of this paper is two-fold. On the algorithmic side, it offers a way to scale up evolutionary search by detecting the underlying structure of the search space. On the biological side, it proves that the distributed representation of the evolutionary fitness function in gene expression can be computed in polynomial-time.

Distributed Clustering Using Collective Principal Component Analysis.
This paper considers distributed clustering of high dimensional heterogeneous data using a distributed  Principal Component Analysis (PCA) technique called the Collective PCA. It presents the Collective  PCA technique that can be used independent of the clustering application. It shows a way to integrate  the Collective PCA with a given o-the-shelf clustering algorithm in order to develop a distributed  clustering technique. It also presents experimental results using dierent test data sets including an  application for web mining.
Computation in Genetic Code%2DLike Transformations.

Distributed and parallel knowledge discovery %28workshop session %2D title only%29.

Collective Principal Component Analysis from Distributed%2C Heterogeneous Data.

The Genetic Code%2DLike Transformations and Their Effect on Learning Functions.

Report from the Workshop on Distributed and Parallel Knowledge Discovery

Collective%2C Hierarchical Clustering from Distributed%2C Heterogeneous Data.
. This paper presents the Collective Hierarchical Clustering  (CHC) algorithm for analyzing distributed, heterogeneous data. This algorithm  first generates local cluster models and then combines them to  generate the global cluster model of the data. The proposed algorithm runs  in O(jSjn  2  ) time, with a O(jSjn) space requirement and O(n) communication  requirement, where n is the number of elements in the data set and  jSj is the number of data sites. This approach shows significant improvement  over naive methods with O(n  2  ) communication costs in the case  that the entire distance matrix is transmitted and O(nm) communication  costs to centralize the data, where m is the total number of features.  A specific implementation based on the single link clustering and results  comparing its performance with that of a centralized clustering algorithm  are presented. An analysis of the algorithm complexity, in terms of overall  computation time and communication requirements, is pres...
Further Experimentations on the Scalability of the GEMGA.
. This paper reports the recent developments of the Gene Expression Messy Genetic Algorithm (GEMGA) research. It presents extensive experimental results for large problems with massive multi-modality, non-uniform scaling, and overlapping sub-problems. All the experimental results corroborate the linear time performance of the GEMGA for a wide range of problems, that can be decomposed into smaller overlapping and non-overlapping sub-problems in the chosen representation. These results further support the scalable performance of the GEMGA. 1 Introduction  The recent past has witnessed a growing interest in designing scalable evolutionary algorithms that inductively detect the decomposable partitions (sometimes called genetic linkage in the Genetic Algorithm (GA) literature[8]) of the optimization problem. Messy genetic algorithms [4, 6, 5, 7, 10, 3], Dependency trees [2], distribution estimation [15, 16], are some examples. For problems that can be decomposed into smaller subproblems in ...
DNA To Protein Transformations and Their Possible Role in Linkage Learning.
This paper first presents an extended perspective of linkage using basic concepts developed in the SEARCH framework (Kargupta, 1995; Kargupta &amp; Goldberg, 1996) and identifies detection of &#034;appropriate&#034; patterns or relations among the search space members as the fundamental and broader objective of linkage learning. It then explores the computational role of gene-expression (DNA!RNA!Protein transformations) in evolutionary search for relations, using an algebraic approach. It offers strong evidence to support the hypothesis that the transformations in gene-expression form fitness invariant symmetry groups over alphabet tuples that may be used to capture patterns or relations among the search space members. 1 Introduction  Intra-cellular expression of genetic information in a living organism plays a critical role in the emergence of different forms of life. Different regions of DNA, the carrier of genetic information, are transcribed in different cells of an organism for producing messen...
Scalable%2C Distributed Data Mining %2D An Agent Architecture.

SEARCH%2C Blackbox Optimization%2C And Sample Complexity.
The SEARCH (Search Envisioned As Relation &amp; Class Hierarchizing) framework developed elsewhere (Kargupta, 1995) offered an alternate perspective toward blackbox optimization (BBO)---optimization in presence of little domain knowledge. The SEARCH framework investigated the conditions essential for transcending the limits of random enumerative search using a framework developed in terms of relations, classes and partial ordering. This paper presents a summary of some of the main results of that work. A closed form bound on the sample complexity in terms of the cardinality of the relation space, class space, desired quality of the solution and the reliability is presented. The two primary lessons of this work are, a BBO (1) must search for appropriate relations and (2) can only solve the so called class of order-k delineable problems in polynomial sample complexity. These results are applicable to any blackbox search algorithms, including evolutionary optimization techniques. 1 Introducti...
The Performance of the Gene Expression Messy Genetic Algorithm On Real Test Functions.

Polynominal Complexity Blackbox Search Lessons From the SEARCH Framework.

The Gene Expression Messy Genetic Algorithm.
This paper introduces the gene expression messy genetic algorithm (GEMGA) a new generation of messy GAs that directly search for relations among the members of the search space. The GEMGA is an O(Ak( 2 q- k)) sam- ple complexity algorithm for the class of order-k deline- able problems [6] (problems that can be solved by considering no higher than order-k relations). The GEMGA is designed based on an alternate perspective of natural evo- lution proposed by the SEARCH framework [6] that em- phasizes the role of gene expression. The GEMGA uses the transcription operator to search for relations. This paper also presents the test results of the GEMGA for large multimodal order-k delineable problems.
Signal%2Dto%2Dnoise
Using a SPICE network model, we simulated the signal and noise of AC-coupled, singlesided,   n  +  on n type Si micro-strip detectors connected to PreMux-128 read-out electronics.  The detector response was studied before and after neutron irradiation with a fluence of  8:3 \Delta 10  13  n=cm  2  . We compared the simulated and experimental signal-to-noise ratios.  1 Introduction  The signal-to-noise ratio (S/N) is the key parameter of Si micro-strip detectors to determine the detector efficiency. The general requirement to achieve good tracking capability in CMS experiment for LHC at CERN is S=N ? 10 [1]. This performance should be maintained over 10 years of operation, by the time the fluence at the hottest part of the system is expected to reach about 1:6 \Delta 10  14  n=cm  2  .  The main component of the noise associated to the measurement of the charge signal in a Si detector is due to the front-end electronics. For a charge sensitive preamplifier, having CMOS input transistor,...
RapidAccurate Optimization of Difficult Problems Using Fast Messy Genetic Algorithms.
Researchers have long sought genetic algorithms (GAs) that can solve difficult search, optimization, and machine learning problems quickly. Despite years of work on simple GAs and their variants it is still unknown how difficult a problem simple GAs can solve, how quickly they can solve it, and with what reliability. More radical design departures than these have been taken, however, and the messy GA (mGA) approach has attempted to solve problems of bounded difficulty quickly and reliably by taking the notion of building-block linkage quite seriously. Early efforts were apparently successful in achieving polynomial convergence on some difficult problems, but the initialization bottleneck that required a large initial population was thought to be the primary obstacle to faster mGA performance. This paper replaces the partially enumerative initialization and selective primordial phase of the original messy GA with  probabilistically complete initialization and a primordial phase that per...
Information Transmission in Genetic Algorithm and Shannon%27s Second Theorem.

Ordering Genetic Algorithms and Deception.
This paper considers deception in the context of ordering genetic algorithms (GAs). Order-four deceptive ordering problems are designed for absolute and relative ordering decoding. Three different crossover operators are used in both absolute and relative ordering problems, and for each combination of crossover operator and coding, the schema survival probability is calculated. Simulation results show that no single crossover operator is adequate to find the globally optimal solution in both absolute and relative ordering problems. This confirms the fundamental GA theory, that the success of a genetic algorithm depends on how well the crossover operator respects the underlying coding of the problem. 1.
System Identification with Evolving Polynomial Networks.
this paper can aid in understanding GA behavior in settings that require diversity. Improved understanding in this area is necessary if GAs are to be actively employed in systems that require cooperating sets of interacting individuals
