From 5fcc08e075f74b416d682b7c16d1ac2d2447dd52 Mon Sep 17 00:00:00 2001 From: Thanh Tung Khuat Date: Wed, 29 Jun 2022 10:21:22 +0700 Subject: [PATCH] Update documentation for several exciting features of GFMMNN --- README.rst | 31 +++++++++++++++++++++++++++++-- docs/user/features.rst | 31 +++++++++++++++++++++++++++++-- index.html | 27 +++++++++++++++++++++++++-- 3 files changed, 83 insertions(+), 6 deletions(-) diff --git a/README.rst b/README.rst index b345a4e..2c77c17 100644 --- a/README.rst +++ b/README.rst @@ -183,8 +183,9 @@ training samples or a subset of both training samples and features. Training sub can be formed by stratified random subsampling, resampling, or class-balanced random subsampling. The final predicted results of an ensemble model are an aggregation of predictions from all base learners based on a majority voting mechanism. An intersting characteristic of hyperbox-based models is resulting -hyperboxes from all base learners can be merged to formulate a single model. This contributes to increasing -the explainability of the estimator while still taking advantage of strong points of ensemble models. +hyperboxes from all base learners or decision trees can be merged to formulate a single model. This +contributes to increasing the explainability of the estimator while still taking advantage of strong points +of ensemble models. Multigranularity learning ~~~~~~~~~~~~~~~~~~~~~~~~~ @@ -194,6 +195,32 @@ abstraction. An attractive characteristic of these classifiers is that they can to other fuzzy min-max models at a low degree of granularity based on reusing the knowledge learned from lower levels of abstraction. +Learning from both labelled and unlabelled data +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +One of the exciting features of learning algorithms for the general fuzzy min-max neural network is the capability of +creating classification boundaries among known classes and clustering data and representing them as hyperboxes in the case that +labels are not available. Unlabelled hyperboxes is then possible to be labelled on the basis of the evidence of next incoming +input samples. As a result, the GFMMNN models have the ability to learn from the mixed labelled and unlabelled datasets in +a native way. + +Ability to directly process missing data +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +Learning algorithms for the general fuzzy min-max neural network supported by the library may classify inputs with missing +data directly without the need for replacing or imputing missing values as in other classifiers. + +Continual learning of new classes in an incremental manner +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +Incremental learning algorithms of hyperbox-based models in the **hyperbox-brain** library can grow and accommodate new +classes of data without retraining the whole classifier. Incremental learning algorithms themselves can generate new hyperboxes +to represent clusters of new data with potentially new class labels both in the middle of normal training procedure and in the +operating time where training has been finished. This property is a key feature for smart life-long learning systems. + +Data editing and pruning approaches +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +By combining the repeated cross-validation methods provided by scikit-learn and hyperbox-based learning algorithms, evidence from +training multiple models can be deployed for identifying which data points from the original training set or the hyperboxes from +the generated multiple models should be retained and those that should be edited out or pruned before further processing. + Scikit-learn compatible estimators ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The estimators in hyperbox-brain is compatible with the well-known scikit-learn toolbox. diff --git a/docs/user/features.rst b/docs/user/features.rst index 52ef471..0365f94 100644 --- a/docs/user/features.rst +++ b/docs/user/features.rst @@ -29,8 +29,9 @@ training samples or a subset of both training samples and features. Training sub can be formed by stratified random subsampling, resampling, or class-balanced random subsampling. The final predicted results of an ensemble model are an aggregation of predictions from all base learners based on a majority voting mechanism. An intersting characteristic of hyperbox-based models is resulting -hyperboxes from all base learners can be merged to formulate a single model. This contributes to increasing -the explainability of the estimator while still taking advantage of strong points of ensemble models. +hyperboxes from all base learners or decision trees can be merged to formulate a single model. This +contributes to increasing the explainability of the estimator while still taking advantage of strong points +of ensemble models. Multigranularity learning ~~~~~~~~~~~~~~~~~~~~~~~~~ @@ -40,6 +41,32 @@ abstraction. An attractive characteristic of these classifiers is that they can to other fuzzy min-max models at a low degree of granularity based on reusing the knowledge learned from lower levels of abstraction. +Learning from both labelled and unlabelled data +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +One of the exciting features of learning algorithms for the general fuzzy min-max neural network is the capability of +creating classification boundaries among known classes and clustering data and representing them as hyperboxes in the case that +labels are not available. Unlabelled hyperboxes is then possible to be labelled on the basis of the evidence of next incoming +input samples. As a result, the GFMMNN models have the ability to learn from the mixed labelled and unlabelled datasets in +a native way. + +Ability to directly process missing data +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +Learning algorithms for the general fuzzy min-max neural network supported by the library may classify inputs with missing +data directly without the need for replacing or imputing missing values as in other classifiers. + +Continual learning of new classes in an incremental manner +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +Incremental learning algorithms of hyperbox-based models in the **hyperbox-brain** library can grow and accommodate new +classes of data without retraining the whole classifier. Incremental learning algorithms themselves can generate new hyperboxes +to represent clusters of new data with potentially new class labels both in the middle of normal training procedure and in the +operating time where training has been finished. This property is a key feature for smart life-long learning systems. + +Data editing and pruning approaches +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +By combining the repeated cross-validation methods provided by scikit-learn and hyperbox-based learning algorithms, evidence from +training multiple models can be deployed for identifying which data points from the original training set or the hyperboxes from +the generated multiple models should be retained and those that should be edited out or pruned before further processing. + Scikit-learn compatible estimators ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The estimators in hyperbox-brain is compatible with the well-known scikit-learn toolbox. diff --git a/index.html b/index.html index f24475b..0ad7ecf 100644 --- a/index.html +++ b/index.html @@ -129,11 +129,34 @@
Ensemble learning
Ensemble models in the hyperbox-brain toolbox build a set of hyperbox-based learners from a subset of training samples or a subset of both training samples and features. Training subsets of base learners can be formed by stratified random subsampling, resampling, or class-balanced random subsampling. The final predicted results of an ensemble model are an aggregation of predictions from all base learners based on a majority voting mechanism. An intersting characteristic of hyperbox-based models is resulting hyperboxes from all base learners can be merged to formulate a single model.

-

+
+

Multigranularity learning
Multi-granularity learning algorithms can construct classifiers from multiresolution hierarchical granular representations using hyperbox fuzzy sets. This algorithm forms a series of granular inferences hierarchically through many levels of abstraction. An attractive characteristic of these classifiers is that they can maintain a high accuracy in comparison to other fuzzy min-max models at a low degree of granularity based on reusing the knowledge learned from lower levels of abstraction.
-

+

+
+
+

+
Learning from both labelled and unlabelled data
+
One of the exciting features of learning algorithms for the general fuzzy min-max neural network(GFMMNN) is the capability of creating classification boundaries among known classes and clustering data and representing them as hyperboxes in the case that labels are not available. Unlabelled hyperboxes is then possible to be labelled on the basis of the evidence of next incoming input samples. As a result, the GFMMNN models have the ability to learn from the mixed labelled and unlabelled datasets in a native way. +
+

+
+
+

+
Ability to directly process missing data
+
Learning algorithms for the general fuzzy min-max neural network supported by the library may classify inputs with missing data directly without the need for replacing or imputing missing values as in other classifiers. +
+

+
+
+

+
Continual learning of new classes in an incremental manner
+
Incremental learning algorithms of hyperbox-based models in the **hyperbox-brain** library can grow and accommodate new classes of data without retraining the whole classifier. Incremental learning algorithms themselves can generate new hyperboxes to represent clusters of new data with potentially new class labels both in the middle of normal training procedure and in the operating time where training has been finished. This property is a key feature for smart life-long learning systems. +
+

+

Scikit-learn compatible estimators
The estimators in hyperbox-brain is compatible with the well-known scikit-learn toolbox.