Abstract: The exponential increase in the volume of medical image database has imposed new challenges to clinical routine in maintaining patient history, diagnosis, treatment and monitoring. With the advent of data mining and machine learning techniques it is possible to automate and/or assist physicians in clinical diagnosis. In this research a medical image classification framework using data mining techniques is proposed. It involves feature extraction, feature selection, feature discretization and classification. In the classification phase, the performance of the traditional kNN k nearest neighbor classifier is improved using a feature weighting scheme and a distance weighted voting instead of simple majority voting. Feature weights are calculated using the interestingness measures used in association rule mining. Experiments on the retinal fundus images show that the proposed framework improves the classification accuracy of traditional kNN from 78.57 % to 92.85 %.
Abstract: Association rules are an important problem in data
mining. Massively increasing volume of data in real life databases
has motivated researchers to design novel and incremental algorithms
for association rules mining. In this paper, we propose an incremental
association rules mining algorithm that integrates shocking
interestingness criterion during the process of building the model. A
new interesting measure called shocking measure is introduced. One
of the main features of the proposed approach is to capture the user
background knowledge, which is monotonically augmented. The
incremental model that reflects the changing data and the user beliefs
is attractive in order to make the over all KDD process more
effective and efficient. We implemented the proposed approach and
experiment it with some public datasets and found the results quite
promising.
Abstract: Rule Discovery is an important technique for mining
knowledge from large databases. Use of objective measures for
discovering interesting rules leads to another data mining problem,
although of reduced complexity. Data mining researchers have
studied subjective measures of interestingness to reduce the volume
of discovered rules to ultimately improve the overall efficiency of
KDD process.
In this paper we study novelty of the discovered rules as a
subjective measure of interestingness. We propose a hybrid approach
based on both objective and subjective measures to quantify novelty
of the discovered rules in terms of their deviations from the known
rules (knowledge). We analyze the types of deviation that can arise
between two rules and categorize the discovered rules according to
the user specified threshold. We implement the proposed framework
and experiment with some public datasets. The experimental results
are promising.
Abstract: Sequential mining methods efficiently discover all frequent sequential patterns included in sequential data. These methods use the support, which is the previous criterion that satisfies the Apriori property, to evaluate the frequency. However, the discovered patterns do not always correspond to the interests of analysts, because the patterns are common and the analysts cannot get new knowledge from the patterns. The paper proposes a new criterion, namely, the sequential interestingness, to discover sequential patterns that are more attractive for the analysts. The paper shows that the criterion satisfies the Apriori property and how the criterion is related to the support. Also, the paper proposes an efficient sequential mining method based on the proposed criterion. Lastly, the paper shows the effectiveness of the proposed method by applying the method to two kinds of sequential data.
Abstract: Term Extraction, a key data preparation step in Text
Mining, extracts the terms, i.e. relevant collocation of words,
attached to specific concepts (e.g. genetic-algorithms and decisiontrees
are terms associated to the concept “Machine Learning" ). In
this paper, the task of extracting interesting collocations is achieved
through a supervised learning algorithm, exploiting a few
collocations manually labelled as interesting/not interesting. From
these examples, the ROGER algorithm learns a numerical function,
inducing some ranking on the collocations. This ranking is optimized
using genetic algorithms, maximizing the trade-off between the false
positive and true positive rates (Area Under the ROC curve). This
approach uses a particular representation for the word collocations,
namely the vector of values corresponding to the standard statistical
interestingness measures attached to this collocation. As this
representation is general (over corpora and natural languages),
generality tests were performed by experimenting the ranking
function learned from an English corpus in Biology, onto a French
corpus of Curriculum Vitae, and vice versa, showing a good
robustness of the approaches compared to the state-of-the-art Support
Vector Machine (SVM).
Abstract: Rule Discovery is an important technique for mining knowledge from large databases. Use of objective measures for discovering interesting rules lead to another data mining problem, although of reduced complexity. Data mining researchers have studied subjective measures of interestingness to reduce the volume of discovered rules to ultimately improve the overall efficiency of KDD process. In this paper we study novelty of the discovered rules as a subjective measure of interestingness. We propose a hybrid approach that uses objective and subjective measures to quantify novelty of the discovered rules in terms of their deviations from the known rules. We analyze the types of deviation that can arise between two rules and categorize the discovered rules according to the user specified threshold. We implement the proposed framework and experiment with some public datasets. The experimental results are quite promising.
Abstract: Knowledge Discovery of Databases (KDD) is the
process of extracting previously unknown but useful and significant
information from large massive volume of databases. Data Mining is
a stage in the entire process of KDD which applies an algorithm to
extract interesting patterns. Usually, such algorithms generate huge
volume of patterns. These patterns have to be evaluated by using
interestingness measures to reflect the user requirements.
Interestingness is defined in different ways, (i) Objective measures
(ii) Subjective measures. Objective measures such as support and
confidence extract meaningful patterns based on the structure of the
patterns, while subjective measures such as unexpectedness and
novelty reflect the user perspective. In this report, we try to brief the
more widely spread and successful subjective measures and propose
a new subjective measure of interestingness, i.e. shocking.
Abstract: Knowledge Discovery in Databases (KDD) is the process of extracting previously unknown, hidden and interesting patterns from a huge amount of data stored in databases. Data mining is a stage of the KDD process that aims at selecting and applying a particular data mining algorithm to extract an interesting and useful knowledge. It is highly expected that data mining methods will find interesting patterns according to some measures, from databases. It is of vital importance to define good measures of interestingness that would allow the system to discover only the useful patterns. Measures of interestingness are divided into objective and subjective measures. Objective measures are those that depend only on the structure of a pattern and which can be quantified by using statistical methods. While, subjective measures depend only on the subjectivity and understandability of the user who examine the patterns. These subjective measures are further divided into actionable, unexpected and novel. The key issues that faces data mining community is how to make actions on the basis of discovered knowledge. For a pattern to be actionable, the user subjectivity is captured by providing his/her background knowledge about domain. Here, we consider the actionability of the discovered knowledge as a measure of interestingness and raise important issues which need to be addressed to discover actionable knowledge.