Search results for: Hiroshi Motoda

Items from 1 to 20 out of 102 results

article

Accurate and efficient detection of critical links in network to minimize information loss

Kazumi Saito, Kouzou Ohara, Masahiro Kimura, Hiroshi Motoda

Journal of Intelligent Information Systems > 2018 > 51 > 2 > 235-255

We address the problem of efficiently detecting critical links in a large network. Critical links are such links that their deletion exerts substantial effects on the network performance such as the average node reachability. We tackle this problem by proposing a new method which consists of three acceleration techniques: redundant-link skipping (RLS), marginal-node pruning (MNP) and burn-out following...

chapter

Discovering Cooperative Structure Among Online Items for Attention Dynamics

Kanji Matsutani, Masahito Kumano, Masahiro Kimura, Kazumi Saito, more

2017 IEEE International Conference on Data Mining Workshops (ICDMW) > 1033 - 1041

2017 IEEE International Conference on Data Mining Workshops (ICDMW)

Social Media allows people to post widely and share the posted online-items. Such items gain their popularity by the amount of attention received. Thus, studies on modeling the arrival process of attention to an individual item have recently attracted a great deal of interest. In this paper, we propose, by combining a Dirichlet process with a Hawkes process in a novel way, a probabilistic model, called...

chapter

Toward the Discovery of First Principle Based Scientific Law Equations

Takashi Washio, Hiroshi Motoda

Lecture Notes in Computer Science > Progress in Discovery Science > 553-564

Conventional work on scientific discovery such as BACON derives empirical law equations from experimental data. In recent years, SDS introducing mathematical admissibility constraints has been proposed to discover first principle based law equations, and it has been further extended to discover law equations from passively observed data. Furthermore, SSF has been proposed to discover the structure...

chapter

Acquiring Expected Influence Curve from Single Diffusion Sequence

Yuya Yoshikawa, Kazumi Saito, Hiroshi Motoda, Kouzou Ohara, more

Lecture Notes in Computer Science > Knowledge Management and Acquisition for Smart Systems and Services > KA Applications in Internet and Mobile Computing > 273-287

We address the problem of estimating the expected influence curves with good accuracy from a single observed information diffusion sequence, for both the asynchronous independent cascade (AsIC) model and the asynchronous linear threshold (AsLT) model. We solve this problem by first learning the model parameters and then estimating the influence curve using the learned model. Since the length of the...

chapter

Basket Analysis for Graph Structured Data

Akihiro Inokuchi, Takashi Washio, Hiroshi Motoda, Kouhei Kumasawa, more

Lecture Notes in Computer Science > Methodologies for Knowledge Discovery and Data Mining > Causal Model and Graph-Based Methods > 421-433

The Basket Analysis derives frequent itemsets and association rules having support and confidence levels greater than their thresholds from massive transaction data. Though some recent research tries to discover wider classes of knowledge on the regularities contained in the data, the regularities in form of the graph structure has not been explored in the field of the Basket Analysis. The work reported...

chapter

Case Generation Method for Constructing an RDR Knowledge Base

Keisei Fujiwara, Tetsuya Yoshida, Hiroshi Motoda, Takashi Washio

Lecture Notes in Computer Science > PRICAI 2002: Trends in Artificial Intelligence > Knowledge Acquisition and Management > 228-237

Ripple Down Rules (RDR) Method is an incremental Knowledge Acquisition (KA) approach that is able to capture human expertise efficiently. The expert’s KA tasks in RDR are 1) to identify the correct class label of each misclassified case and 2) to select important attributes that distinguish the misclassified case from the previous correctly classified case. The latter task is more difficult than the...

chapter

Speeding Up Bipartite Graph Visualization Method

Takayasu Fushimi, Yamato Kubota, Kazumi Saito, Masahiro Kimura, more

Lecture Notes in Computer Science > AI 2011: Advances in Artificial Intelligence > Session 9: Cognitive Modeling and Simulation Technology > 697-706

We address the problem of visualizing structure of bipartite graphs such as relations between pairs of objects and their multi-labeled categories. For this task, the existing spherical embedding method, as well as the other standard graph embedding methods, can be used. However, these existing methods either produce poor visualization results or require extremely large computation time to obtain the...

chapter

A new algorithm for automatic configuration of Hidden Markov Models

Makoto Iwayama, Nitin Indurkhya, Hiroshi Motoda

Lecture Notes in Computer Science > Algorithmic Learning Theory > 237-250

Hidden Markov Models (HMM) (i.e. doubly stochastic probabilistic networks) have been widely used in analyzing time-series data such as those obtained from speech and molecular biology. A crucial issue in modeling time-series data using HMM, is the problem of determining the appropriate model architecture: the number of states and the links between the states. While current HMM training procedures...

chapter

A Data Pre-processing Method Using Association Rules of Attributes for Improving Decision Tree

Masahiro Terabe, Osamu Katai, Tetsuo Sawaragi, Takashi Washio, more

Lecture Notes in Computer Science > Methodologies for Knowledge Discovery and Data Mining > Feature Selection and Generation > 143-147

One of the approaches to generate a good decision tree is preprocessing the data to improve its description. There are many researches on data pre-processing such as attributes generation and attributes selection methods. However, most of them are based on logic programming so that it takes much run time. Additionally, some of them need a priori knowledge. These are disadvantage for the data mining...

chapter

Communicability Criteria of Law Equations Discovery

Takashi Washio, Hiroshi Motoda

Lecture Notes in Computer Science > Computational Discovery of Scientific Knowledge > I Equation Discovery and Dynamic Systems Identification > 98-119

The “laws” in science are not the relations established by only the objective features of the nature. They have to be consistent with the assumptions and the operations commonly used in the study of scientists identifying these relations. Upon this consistency, they become communicable among the scientists. The objectives of this literature are to discuss a mathematical foundation of the communicability...

chapter

What Can We Do with Graph-Structured Data? – A Data Mining Perspective

Hiroshi Motoda

Lecture Notes in Computer Science > AI 2006: Advances in Artificial Intelligence > Invited Talks > 1-2

Recent advancement of data mining techniques has made it possible to mine from complex structured data. Since structure is represented by proper relations and a graph can easily represent relations, knowledge discovery from graph-structured data (graph mining) poses a general problem for mining from structured data. Some examples amenable to graph mining are finding functional components from their...

chapter

Analysis of Hepatitis Dataset by Decision Tree Based on Graph-Based Induction

Warodom Geamsakul, Takashi Matsuda, Tetsuya Yoshida, Kouzou Ohara, more

Lecture Notes in Computer Science > New Frontiers in Artificial Intelligence > 17th Annual Conference of the Japanese Society for Artificial Intelligence (JSAI 2003) – Award-Winning Papers > 5-28

A machine learning technique called Graph-Based Induction (GBI) efficiently extracts typical patterns from graph-structured data by stepwise pair expansion (pairwise chunking). It is very efficient because of its greedy search. We have expanded GBI to construct a decision tree that can handle graph-structured data. DT-GBI constructs a decision tree while simultaneously constructing attributes for...

chapter

Feature Selection Using Consistency Measure

Manoranjan Dash, Huan Liu, Hiroshi Motoda

Lecture Notes in Computer Science > Discovery Science > Posters > 319-320

Feature selection methods search for an “optimal” subset of features. Many methods exist. We evaluate consistency measure along with different search techniques applied in the literature and suggest a guideline of its use.

chapter

Active Feature Selection Using Classes

Huan Liu, Lei Yu, Manoranjan Dash, Hiroshi Motoda

Lecture Notes in Computer Science > Advances in Knowledge Discovery and Data Mining > Feature Selection > 474-485

Feature selection is frequently used in data pre-processing for data mining. When the training data set is too large, sampling is commonly used to overcome the difficulty. This work investigates the applicability of active sampling in feature selection in a filter model setting. Our objective is to partition data by taking advantage of class information so as to achieve the same or better performance...

chapter

Consistency Based Feature Selection

Manoranjan Dash, Huan Liu, Hiroshi Motoda

Lecture Notes in Computer Science > Knowledge Discovery and Data Mining. Current Issues and New Applications > Feature Selection and Transformation > 98-109

Feature selection is an effective technique in dealing with dimensionality reduction for classification task, a main component of data mining. It searches for an “optimal” subset of features. The search strategies under consideration are one of the three: complete, heuristic, and probabilistic. Existing algorithms adopt various measures to evaluate the goodness of feature subsets. This work focuses...

chapter

Workshop on Active Mining (AM-2004)

Masayuki Numao, Takahira Yamaguchi, Shusaku Tsumoto, Hiroshi Motoda

Lecture Notes in Computer Science > New Frontiers in Artificial Intelligence > Workshop on Active Mining > 463-463

The workshop on Active Mining (AM-2004) was held on June 1, 2004 at Ishikawa Kousei Nenkin Kaikan in Kanazawa City, Japan, as a part of the Eighteenth Annual Conference of the Japanese Society for Artificial Intelligence (JSAI-2004). This is the third workshop that focuses on Active Mining; the first one was held on December 9, 2002, as a part of the Second IEEE International Conference on Data Mining...

chapter

Extracting Diagnostic Knowledge from Hepatitis Dataset by Decision Tree Graph-Based Induction

Warodom Geamsakul, Tetsuya Yoshida, Kouzou Ohara, Hiroshi Motoda, more

Lecture Notes in Computer Science > Active Mining > Active Data Mining > 126-151

We have proposed a method called Decision Tree Graph-Based Induction (DT-GBI), which constructs a classifier (decision tree) for graph-structured data while simultaneously constructing attributes for classification. Graph-Based Induction (GBI) is utilized in DT-GBI for efficiently extracting typical patterns from graph-structured data by stepwise pair expansion (pairwise chunking). Attributes, i.e...

chapter

Graph-Based Induction for General Graph Structured Data

Takashi Matsuda, Tadashi Horiuchi, Hiroshi Motoda, Takashi Washio, more

Lecture Notes in Computer Science > Discovery Science > Posters > 340-342

chapter

Discovering Influential Nodes for SIS Models in Social Networks

Kazumi Saito, Masahiro Kimura, Hiroshi Motoda

Lecture Notes in Computer Science > Discovery Science > 302-316

We address the problem of efficiently discovering the influential nodes in a social network under the susceptible/infected/susceptible (SIS) model, a diffusion model where nodes are allowed to be activated multiple times. The computational complexity drastically increases because of this multiple activation property. We solve this problem by constructing a layered graph from the original social network...

chapter

Finding Relation between PageRank and Voter Model

Takayasu Fushimi, Kazumi Saito, Masahiro Kimura, Hiroshi Motoda, more

Lecture Notes in Computer Science > Knowledge Management and Acquisition for Smart Systems and Services > KA Applications in Internet and Mobile Computing > 208-222

Estimating influence of a node is an important problem in social network analyses. We address this problem in a particular class of model for opinion propagation in which a node adopts its opinion based on not only its direct neighbors but also the average opinion share over the whole network, which we call an extended Voter Model with uniform adoption (VM). We found a similarity of this model with...

Publication date

Set your own date range

Content availability

Available (96)
None (6)

Publication type

book (76)
article (26)

INFONA - science communication portal

Search results for: Hiroshi Motoda

Accurate and efficient detection of critical links in network to minimize information loss

Discovering Cooperative Structure Among Online Items for Attention Dynamics

Toward the Discovery of First Principle Based Scientific Law Equations

Acquiring Expected Influence Curve from Single Diffusion Sequence

Basket Analysis for Graph Structured Data

Case Generation Method for Constructing an RDR Knowledge Base

Speeding Up Bipartite Graph Visualization Method

A new algorithm for automatic configuration of Hidden Markov Models

A Data Pre-processing Method Using Association Rules of Attributes for Improving Decision Tree

Communicability Criteria of Law Equations Discovery

What Can We Do with Graph-Structured Data? – A Data Mining Perspective

Analysis of Hepatitis Dataset by Decision Tree Based on Graph-Based Induction

Feature Selection Using Consistency Measure

Active Feature Selection Using Classes

Consistency Based Feature Selection

Workshop on Active Mining (AM-2004)

Extracting Diagnostic Knowledge from Hepatitis Dataset by Decision Tree Graph-Based Induction

Graph-Based Induction for General Graph Structured Data

Discovering Influential Nodes for SIS Models in Social Networks

Finding Relation between PageRank and Voter Model

Filter options

Publication date

Content availability

Publication type

Keywords

Data set

Journal

INFONA - science communication portal

Search results for: Hiroshi Motoda

Add recipient

Sending message cancelled

Are you sure you want to cancel sending this message?

Send message

Filter options

Publication date

Date range setting

Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.

Content availability

Publication type

Keywords

Data set

Journal

Reporting an error / abuse

Sending the report failed

Accessibility options