Piecewise linear models (PLMs) have been widely used in many enterprise machine learning problems, which assign linear experts to individual partitions on feature spaces and express whole models as patches of local experts. This paper addresses simultaneous model selection issues of PLMs; partition structure determination and feature selection of individual experts. Our contributions are mainly three-fold. First, we extend factorized asymptotic Bayesian (FAB) inference for hierarchical mixtures of experts (probabilistic PLMs). FAB inference offers penalty terms w.r.t. partition and expert complexities, and enable us to resolve the model selection issue. Second, we propose posterior optimization which significantly improves predictive accuracy. Roughly speaking, our new posterior optimization mitigates accuracy degradation due to a gap between marginal log-likelihood maximization and predictive accuracy. Third, we present an application of energy demand forecasting as well as benchmark comparisons. The experiments show our capability of acquiring compact and highly-accurate models.

It had been believed in the conventional practice that the risk of a bank going bankrupt is lessened in a straightforward manner by transferring the risk of loan defaults. But the failure of American International Group in 2008 posed a more complex aspect of financial contagion. This study presents an extension of the asset network systemic risk model (ANWSER) to investigate whether credit default swaps mitigate or intensify the severity of financial contagion. A protection buyer bank transfers the risk of every possible debtor bank default to protection seller banks. The empirical distribution of the number of bank bankruptcies is obtained with the extended model. Systemic capital buffer ratio is calculated from the distribution. The ratio quantifies the effective loss absorbency capability of the entire financial system to force back financial contagion. The key finding is that the leverage ratio is a good estimate of a systemic capital buffer ratio as the backstop of a financial system. The risk transfer from small and medium banks to big banks in an interbank network does not mitigate the severity of financial contagion.

Most time-varying data in our daily life is multi-variate. Moreover, most of such time-varying data contains both numeric and categorical values. It is often meaningful to visualize both of them as they are often correlated. We aim to visualize every value in such time-varying data in a single display space so that we can discover interesting relationships among the values of the time-varying data. This paper presents a heat map-based time-varying data visualization technique which displays both numeric and categorical values in a single display space. The technique assigns time to the horizontal axis of the display space, and vertically arranges the series of colored belts corresponding to the time-sequence values. It generates one belt for a numeric value, and multiple belts for a categorical value. It clusters the belts according to the similarity of color sequences, and re-arranges the belts based on the clustering result. This paper shows an example of the visualization result applying a time-varying multi-variate marketing dataset.

This study presents an ANWSER model (asset network systemic risk model) to quantify the risk of financial contagion which manifests itself in a financial crisis. The transmission of financial distress is governed by a heterogeneous interbank credit network and an investment portfolio of banks. Bankruptcy reproductive ratio of a financial system is computed as a function of the diversity and risk exposure of an investment portfolio of banks, and the denseness and concentration of a heterogeneous interbank credit network. An analytic solution of the bankruptcy reproductive ratio for a small financial system is derived and a numerical solution for a large financial system is obtained. For a large financial system, large diversity among banks in the investment portfolio makes financial contagion more damaging on the average. But large diversity is essentially effective in eliminating the risk of financial contagion in the worst case of financial crisis scenarios. A bank-unique specialization portfolio is more suitable than a uniform diversification portfolio and a system-wide specialization portfolio in strengthening the robustness of a financial system.

Multidimensional data visualization is an important research topic that has been receiving increasing attention. Several techniques that use parallel coordinate plots have been proposed to represent all dimensions of data in a single display space. In addition, several other techniques that apply scatter plot matrices have been proposed to represent multidimensional data as a collection of low-dimensional data visualization spaces. Typically, when using the latter approach it is easier to understand relations among particular dimensions, but it is often difficult to observe relations between dimensions separated into different visualization spaces. This paper presents a framework for displaying an arrangement of low-dimensional data visualization spaces that are generated from high-dimensional datasets. Our proposed technique first divides the dimensions of the input datasets into groups of lower dimensions based on their correlations or other relationships. If the groups of lower dimensions can be visualized in independent rectangular spaces, our technique packs the set of low-dimensional data visualizations into a single display space. Because our technique places relevant low-dimensions closer together in the display space, it is easier to visually compare relevant sets of low-dimensional data visualizations. In this paper, we describe in detail how we implement our framework using parallel coordinate plots, and present several results demonstrating its effectiveness.

This paper proposes an online mixture modeling methodology in which individual components can have different marginal distributions and dependency structures. Mixture models have been widely studied and applied to various application areas, including density estimation, fraud/failure detection, image segmentation, etc. Previous research has been almost exclusively focused on mixture models having components of a single type (e.g., a Gaussian mixture model.) However, recent growing needs for complicated data modeling necessitate the use of more flexible mixture models (e.g., a mixture of a lognormal distribution for medical costs and a Gaussian distribution for blood pressure, for medical analytics.) Our key ideas include: 1) separating marginal distributions and their dependencies using copulas and 2) online extension of a recently-developed "expectation minimization of description length," which enable us to efficiently learn types of both marginal distributions and copulas as well as their parameters. The proposed method provides not only good performance in applications, but also scalable, automatic model selection, which greatly reduces the intensive modeling costs in data mining processes. We show that the proposed method outperforms state-of-the-art methods in application to density estimation and to anomaly detection.

In this paper, we propose a framework to make the text clustering process, as a whole, efficient. In a real text clustering task, an analyst usually has some expectation on the results in mind. However, a single run of a clustering algorithm on the preprocessed data would not satisfy the expectation. Then the analyst faces labor-intensive trials for improving the results that involve repetitive feature refinement and parameter tuning. We develop the Iterative and Interactive Metric Learning System (IIMLS) for addressing the challenge. Specifically, IIMLS allows analysts to input feedback on a current clustering result. Given the feedback, IIMLS optimizes metric in the feature space so that the clustering algorithm applied with the refined metric would reflect the feedback. As a byproduct, learned metric may be used for a similar dataset. Illustrative examples on a real-world dataset show IIMLS can dramatically improve efficiency of a text clustering task. The learned ¿knowledge¿, or the metric, is visualized for gaining insights of the optimized feature metric.

We propose a new text mining system which extracts characteristic contents from given documents. We define Key semantics as characteristic sub-structures of syntactic dependencies in the given documents, and consider the following three tasks in this paper: 1)Key semantics extraction: extracting characteristic syntactic dependency structures not only as ordered trees but also as unordered trees and free trees, 2)Redundancy reduction: from the result of extraction, deleting redundant dependency structures such as sub-structures or equivalent structures of the others, and 3)Phrase/sentence reconstruction: generating a phrase or sentence in a natural language corresponding to the extracted structure.Our system is a combination of natural language processing techniques and tree mining techniques. The system consists of the following five units: 1) syntactic dependency analysis unit, 2) input filters, 3) characteristic ordered subtree extraction unit, 4) output filters, and 5) phrase/sentence reconstruction unit. Although ordered trees are extracted in the third unit, the overall behavior of the system can be switched into the extraction of ordered trees, unordered trees, or free trees depending on which of the input filters is/are applied in the second step. The output filters delete redundant trees from the extraction result for efficient knowledge discovery. Finally, phrases or sentences corresponding to the extracted subtrees are reconstructed by utilizing the input documents.We demonstrate the validity of our system by showing experimental results using real data collected at a help desk and TDT pilot corpus.

In a wide range of business areas dealing with text data streams, including CRM, knowledge management, and Web monitoring services, it is an important issue to discover topic trends and analyze their dynamics in real-time. Specifically we consider the following three tasks in topic trend analysis: 1)Topic Structure Identification; identifying what kinds of main topics exist and how important they are, 2)Topic Emergence Detection; detecting the emergence of a new topic and recognizing how it grows, 3)Topic Characterization; identifying the characteristics for each of main topics. For real topic analysis systems, we may require that these three tasks be performed in an on-line fashion rather than in a retrospective way, and be dealt with in a single framework. This paper proposes a new topic analysis framework which satisfies this requirement from a unifying viewpoint that a topic structure is modeled using a finite mixture model and that any change of a topic trend is tracked by learning the finite mixture model dynamically. In this framework we propose the usage of a time-stamp based discounting learning algorithm in order to realize real-time topic structure identification. This enables tracking the topic structure adaptively by forgetting out-of-date statistics. Further we apply the theory of dynamic model selection to detecting changes of main components in the finite mixture model in order to realize topic emergence detection. We demonstrate the effectiveness of our framework using real data collected at a help desk to show that we are able to track dynamics of topic trends in a timely fashion.

We consider the situation where a number of agents are distributed and each of them collects a data sequence generated according to an unknown probability distribution. Here each of the distributions is specified by common parameters and individual parameters e.g., a normal distribution with an identical mean and a different variance. Here we introduce a notion of an information consortium, which is a framework where the agents cannot show raw data to one another, but they like to enjoy significant information gain for estimating the respective distributions. Such an information consortium has recently received much interest in a broad range of areas including financial risk management, ubiquitous network mining, etc. In this paper we are concerned with the following three issues: 1) how to design a collaborative strategy for agents to estimate the respective distributions in the information consortium, 2) characterizing when each agent has a benefit in terms of information gain for estimating its distribution or information loss for predicting future data, and 3) charracterizing how much benefit each agent obtains. In this paper we yield a statistical formulation of information consortia and solve all of the above three problems for a general form of probability distributions. Specifically we propose a basic strategy for cooperative estimation and derive a necessary and sufficient condition for each agent to have a significant benefit.

Knowing the reputations of your own and/or competitors' products is important for marketing and customer relationship management. It is, however, very costly to collect and analyze survey data manually. This paper presents a new framework for mining product reputations on the Internet. It automatically collects people's opinions about target products from Web pages, and it uses text mining techniques to obtain the reputations of those products. On the basis of human-test samples, we generate in advance syntactic and linguistic rules to determine whether any given statement is an opinion or not, as well as whether such any opinion is positive or negative in nature. We first collect statements regarding target products using a general search engine, and then, using the rules, extract opinions