E-Moderators are \"the new generation of teachers and trainers who work with learners online\" (p. viii) using Computer-Mediated Conferencing (CMC) as a learning tool, regardless of the subject they are teaching. They are the focus of E-Moderating, a recent book which provides both a theoretical framework for developing online learning using CMC (part one), and a wealth of practical advice (part two). The book is supported by a Web site. The author, Gilly Salmon, a distance education specialist with the Open University Business School in the UK, provides a five-step model of effective online education, along with copious examples of how the model relates to real-life online learning contexts. Salmon proposes that, by basing learning on a constructivist model, it is e-moderators that can make the difference in online education as they convene, direct, summarize, and archive synchronous and asynchronous discussions.

/pdf/e-moderating-the-key-to-teaching-and-learning-online-149mr9c3qc.pdf

E-Moderating: The Key to Teaching and Learning Online

We survey agglomerative hierarchical clustering algorithms and discuss efficient implementations that are available in R and other software environments. We look at hierarchical self-organizing maps and mixture models. We review grid-based clustering, focusing on hierarchical density-based approaches. Finally, we describe a recently developed very efficient (linear time) hierarchical clustering algorithm, which can also be viewed as a hierarchical grid-based algorithm. This review adds to the earlier version, Murtagh F, Contreras P. Algorithms for hierarchical clustering: an overview, Wiley Interdiscip Rev: Data Mining Knowl Discov 2012, 2, 86–97. WIREs Data Mining Knowl Discov 2017, 7:e1219. doi: 10.1002/widm.1219

This article is categorized under: 

Algorithmic Development > Hierarchies and Trees
Technologies > Classification
Technologies > Structure Discovery and Clustering

Algorithms for hierarchical clustering: an overview

Neural networks have become an increasingly popular solution for network intrusion detection systems (NIDS). Their capability of learning complex patterns and behaviors make them a suitable solution for differentiating between normal traffic and network attacks. However, a drawback of neural networks is the amount of resources needed to train them. Many network gateways and routers devices, which could potentially host an NIDS, simply do not have the memory or processing power to train and sometimes even execute such models. More importantly, the existing neural network solutions are trained in a supervised manner. Meaning that an expert must label the network traffic and update the model manually from time to time. 
In this paper, we present Kitsune: a plug and play NIDS which can learn to detect attacks on the local network, without supervision, and in an efficient online manner. Kitsune's core algorithm (KitNET) uses an ensemble of neural networks called autoencoders to collectively differentiate between normal and abnormal traffic patterns. KitNET is supported by a feature extraction framework which efficiently tracks the patterns of every network channel. Our evaluations show that Kitsune can detect various attacks with a performance comparable to offline anomaly detectors, even on a Raspberry PI. This demonstrates that Kitsune can be a practical and economic NIDS.

/pdf/kitsune-an-ensemble-of-autoencoders-for-online-network-1qeoj96vls.pdf

Kitsune: An Ensemble of Autoencoders for Online Network Intrusion Detection.

High dimension, low sample size data are emerging in various areas of science. We find a common structure underlying many such data sets by using a non-standard type of asymptotics: the dimension tends to ∞ while the sample size is fixed. Our analysis shows a tendency for the data to lie deterministically at the vertices of a regular simplex. Essentially all the randomness in the data appears only as a random rotation of this simplex. This geometric representation is used to obtain several new statistical insights. Copyright 2005 Royal Statistical Society.

Geometric representation of high dimension, low sample size data

Artificial intelligence and digital pathology: Challenges and opportunities

We survey agglomerative hierarchical clustering algorithms and discuss efficient implementations that are available in R and other software environments. We look at hierarchical self-organizing maps, and mixture models. We review grid-based clustering, focusing on hierarchical density-based approaches. Finally we describe a recently developed very efficient (linear time) hierarchical clustering algorithm, which can also be viewed as a hierarchical grid-based algorithm.

Methods of Hierarchical Clustering

Coding of data, usually upstream of data analysis, has crucial implications for the data analysis results. By modifying the data coding—through use of less than full precision in data values—we can aid appreciably the effectiveness and efficiency of the hierarchical clustering. In our first application, this is used to lessen the quantity of data to be hierarchically clustered. The approach is a hybrid one, based on hashing and on the Ward minimum variance agglomerative criterion. In our second application, we derive a hierarchical clustering from relationships between sets of observations, rather than the traditional use of relationships between the observations themselves. This second application uses embedding in a Baire space, or longest common prefix ultrametric space. We compare this second approach, which is of $O(n \log n)$ complexity, to k-means.

/pdf/hierarchical-clustering-of-massive-high-dimensional-data-1hxa7hj39v.pdf

Hierarchical Clustering of Massive, High Dimensional Data Sets by Exploiting Ultrametric Embedding

The Baire metric induces an ultrametric on a dataset and is of linear computational complexity, contrasted with the standard quadratic time agglomerative hierarchical clustering algorithm. In this work we evaluate empirically this new approach to hierarchical clustering. We compare hierarchical clustering based on the Baire metric with (i) agglomerative hierarchical clustering, in terms of algorithm properties; (ii) generalized ultrametrics, in terms of definition; and (iii) fast clustering through k-means partitioning, in terms of quality of results. For the latter, we carry out an in depth astronomical study. We apply the Baire distance to spectrometric and photometric redshifts from the Sloan Digital Sky Survey using, in this work, about half a million astronomical objects. We want to know how well the (more costly to determine) spectrometric redshifts can predict the (more easily obtained) photometric redshifts, i.e. we seek to regress the spectrometric on the photometric redshifts, and we use clusterwise regression for this.

/pdf/fast-linear-time-hierarchical-clustering-using-the-baire-3o5te5sn0m.pdf

Fast, Linear Time Hierarchical Clustering using the Baire Metric

Identification of inexpensive and technically simple immunological tests useful in predicting the progression to AIDS in human immunodeficiency virus (HIV)-infected patients would be especially welcome in developing countries, in which 80% of HIV-infected patients reside and health budgets are low. In the current study, we evaluated CD4+ and total lymphocyte counts and the concentrations in serum of beta 2-microglobulin, p24 antigen, and immunoglobulin A (IgA) as predictors of disease progression in 74 Panamanian HIV-positive patients and 50 HIV-negative healthy individuals. Total lymphocyte and CD4(+)-cell counts for AIDS patients (1,451 +/- 811 cells/microliters, P 0.05, and 784 +/- 475 cells/microliters, P 0.05) for concentrations of p24 antigen between asymptomatic infected patients (29 +/- 13 pg/ml) and AIDS patients (40 +/- 23 pg/ml) were observed. Total lymphocyte counts of 1,750 cells/microliters or less, CD4 counts of 200 cells/microliters or less, beta 2-microglobulin concentrations in serum of 4 mg/liter or higher, concentrations of IgA in serum of 450 mg/dl or higher, and the presence in serum of p24 antigen were correlated with elevated risks for developing AIDS. Monitoring both total lymphocytes and beta 2-microglobulin identified 91% of the AIDS patients; these assays may allow reductions in the annual number of CD4(+)-cell evaluations and the costs associated with monitoring both total lymphocytes and beta 2-microglobulin identified 91% of the AIDS patients; these assays may allow reductions in the annual number of CD4(+)-cell evaluations and the costs associated with monitoring the immune status of HIV-positive patients.

Pedro Contreras

Papers

Algorithms for hierarchical clustering: an overview

Methods of Hierarchical Clustering

Hierarchical Clustering of Massive, High Dimensional Data Sets by Exploiting Ultrametric Embedding

Fast, Linear Time Hierarchical Clustering using the Baire Metric

Immunological markers of disease progression in patients infected with the human immunodeficiency virus.