Takashi Matsuda

Extension of Graph-Based Induction for General Graph Structured Data

A machine learning technique called Graph-Based Induction (GBI) efficiently extracts typical patterns from directed graph data by stepwise pair expansion (pairwise chunking). In this paper, we expand the capability of the Graph-Based Induction to handle not only tree structured data but also multi-inputs/outputs nodes and loop structure (including a self-loop) which cannot be treated in the conventional way. The method is verified to work as expected using artificially generated data and we evaluated experimentally the computation time of the implemented program. We, further, show the effectiveness of our approach by applying it to two kinds of the real-world data: World Wide Web browsing data and DNA sequence data.

Mining Patterns from Structured Data by Beam-Wise Graph-Based Induction

A machine learning technique called Graph-Based Induction (GBI) extracts typical patterns from graph data by stepwise pair expansion (pairwise chunking). Because of its greedy search strategy, it is very efficient but suffers from incompleteness of search. Improvement is made on its search capability without imposing much computational complexity by 1) incorporating a beam search, 2) using a different evaluation function to extract patterns that are more discriminatory than those simply occurring frequently, and 3) adopting canonical labeling to enumerate identical patterns accurately. This new algorithm, now called Beam-wise GBI, B-GBI for short, was tested against a small DNA dataset from UCI repository and shown successful in extracting discriminatory substructures.

Classifier construction by graph-based induction for graph-structured data

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. Meanwhile, a decision tree is an effective means of data classification from which rules that are easy to understand can be obtained. However, a decision tree could not be produced for the data which is not explicitly expressed with attribute-value pairs. In this paper, we proposes a method of constructing a classifier (decision tree) for graph-structured data by GBI. In our approach attributes, namely substructures useful for classification task, are constructed by GBI on the fly while constructing a decision tree. We call this technique Decision Tree - Graph-Based Induction (DT-GBI). DT-GBI was tested against a DNA dataset from UCI repository. Since DNA data is a sequence of symbols, representing each sequence by attribute-value pairs by simply assigning these symbols to the values of ordered attributes does not make sense. The sequences were transformed into graph-structured data and the attributes (substructures) were extracted by GBI to construct a decision tree. Effect of adjusting the number of times to run GBI at each node of a decision tree is evaluated with respect to the predictive accuracy. The results indicate the effectiveness of DT-GBI for constructing a classifier for graph-structured data.

Graph-based induction and its applications

Abstract A machine learning technique called Graph-based induction (GBI) efficiently extracts typical patterns from graph data by stepwise pair expansion (pairwise chunking). In this paper, we introduce GBI for general graph structured data, which can handle directed/undirected, colored/uncolored graphs with/without (self) loop and with colored/uncolored links. We show that its time complexity is almost linear with the size of graph. We, further, show that GBI can effectively be applied to the extraction of typical patterns from DNA sequence data and organochlorine compound data from which are to be generated classification rules, and that GBI also works as a feature construction component for other machine learning tools.

Performance Evaluation of Decision Tree Graph-Based Induction

A machine learning technique called Decision tree Graph-Based Induction (DT-GBI) constructs a classifier (decision tree) for graph-structured data, which are usually not explicitly expressed with attribute-value pairs. Substructures (patterns) are extracted at each node of a decision tree by stepwise pair expansion (pairwise chunking) in GBI and they are used as attributes for testing. DT-GBI is efficient since GBI is used to extract patterns by greedy search and the obtained result (decision tree) is easy to understand. However, experiments against a DNA dataset from UCI repository revealed that the predictive accuracy of the classifier constructed by DT-GBI was not high enough compared with other approaches. Improvement is made on its predictive accuracy and the performance evaluation of the improved DT-GBI is reported against the DNA dataset. The predictive accuracy of a decision tree is affected by which attributes (patterns) are used and how it is constructed. To extract good enough discriminative patterns, search capability is enhanced by incorporating a beam search into the pairwise chunking within the greedy search framework. Pessimistic pruning is incorporated to avoid overfitting to the training data. Experiments using a DNA dataset were conducted to see the effect of the beam width, the number of chunking at each node of a decision tree, and the pruning. The results indicate that DT-GBI that does not use any prior domain knowledge can construct a decision tree that is comparable to other classifiers constructed using the domain knowledge.

Knowledge Discovery from Structured Data by Beam-Wise Graph-Based Induction

A machine learning technique called Graph-Based Induction (GBI) extracts typical patterns from graph data by stepwise pair expansion (pairwise chunking). Because of its greedy search strategy, it is very efficient but suffers from incompleteness of search. We improved its search capability without imposing much computational complexity by incorporating the idea of beam search. Additional improvement is made to extract patterns that are more discriminative than those simply occurring frequently, and to enumerate identical patterns accurately based on the notion of canonical labeling. This new algorithm was implemented (now called Beam-wise GBI, B-GBI for short) and tested against a DNA data set from UCI repository. Since DNA data is a sequence of symbols, representing each sequence by attribute-value pairs by simply assigning these symbols to the values of ordered attributes does not make sense. By transforming the sequence into a graph structure and running B-GBI it is possible to extract discriminative substructures. These can be new attributes for a classification problem. Effect of beam width on the number of discovered attributes and predictive accuracy was evaluated, together with extracted characteristic subsequences, and the results indicate the effectiveness of B-GBI.

Graph-Based Induction for General Graph Structured Data and Its Application to Chemical Compound Data

Most of the relations are represented by a graph structure, e.g., chemical bonding, Web browsing record, DNA sequence, Inference pattern (program trace), to name a few. Thus, efficiently finding characteristic substructures in a graph will be a useful technique in many important KDD/ML applications. However, graph pattern matching is a hard problem. We propose a machine learning technique called Graph-Based Induction (GBI) that efficiently extracts typical patterns from graph data in an approximate manner by stepwise pair expansion (pairwise chunking). It can handle general graph structured data, i.e., directed/ undirected, colored/uncolored graphs with/without (self) loop and with colored/uncolored links. We show that its time complexity is almost linear with the size of graph. We, further, show that GBI can effectively be applied to the extraction of typical patterns from chemical compound data from which to generate classification rules, and that GBI also works as a feature construction component for other machine learning tools.