Incremental learning with partial instance memory

Maloof, Marcus A.; Michalski, Ryszard S.

doi:10.1016/j.artint.2003.04.001

Cited by 102 publications

(65 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A future work item of ours will be to limit the number of training examples to a certain time window and use a technique called partial memory [25] to reduce the number of training examples.…”

Section: Machine Learning Techniquesmentioning

confidence: 99%

“…The first requirement yields supervised machine learning techniques, that is techniques that can learn from training examples. The requirement for an understandable classifier further limits the range of techniques to symbolic learning techniques, that is techniques that present the learned concept in a human readable form (e.g., predictive rules, decision trees, Prolog clauses) [25].…”

Section: Machine Learning Techniquesmentioning

confidence: 99%

“…Ours is an incremental learning task which is best solved with an incremental learning technique, but can also be solved with a batch learner [14]. As we did not have a working implementation of a purely incremental rule learner (e.g., AQ11 [29], AQ11-PM [25]) we decided to use a "batchincremental" approach.…”

Section: Machine Learning Techniquesmentioning

confidence: 99%

“…In intrusion detection, machine learning has been used primarily to build systems that classify network connections (e.g., 1999 KDD CUP [15]) or system call sequences (e.g., [25]) into one of several predefined classes.…”

Section: Related Workmentioning

confidence: 99%

See 3 more Smart Citations

Using Adaptive Alert Classification to Reduce False Positives in Intrusion Detection

Pietraszek

2004

Lecture Notes in Computer Science

144

View full text Add to dashboard Cite

Abstract. Intrusion Detection Systems (IDSs) are used to monitor computer systems for signs of security violations. Having detected such signs, IDSs trigger alerts to report them. These alerts are presented to a human analyst, who evaluates them and initiates an adequate response. In practice, IDSs have been observed to trigger thousands of alerts per day, most of which are false positives (i.e., alerts mistakenly triggered by benign events). This makes it extremely difficult for the analyst to correctly identify the true positives (i.e., alerts related to attacks). In this paper we describe ALAC, the Adaptive Learner for Alert Classification, which is a novel system for reducing false positives in intrusion detection. The system supports the human analyst by classifying alerts into true positives and false positives. The knowledge of how to classify alerts is learned adaptively by observing the analyst. Moreover, ALAC can be configured to process autonomously alerts that have been classified with high confidence. For example, ALAC may discard alerts that were classified with high confidence as false positive. That way, ALAC effectively reduces the analyst's workload. We describe a prototype implementation of ALAC and the choice of a suitable machine learning technique. Moreover, we experimentally validate ALAC and show how it facilitates the analyst's work.

show abstract

Section: Machine Learning Techniquesmentioning

confidence: 99%

Section: Machine Learning Techniquesmentioning

confidence: 99%

Section: Machine Learning Techniquesmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Using Adaptive Alert Classification to Reduce False Positives in Intrusion Detection

Pietraszek

2004

Lecture Notes in Computer Science

144

View full text Add to dashboard Cite

show abstract

“…In this direction, several attempts have been made, e.g. learning of words [12], language acquisition [15], and incremental learning [9]. However, none of the models proposed in the literature are able to learn, adapt and generalize patterns quickly and effortlessly to recognize new variants of known words and novel words [5] [16].…”

Section: Introductionmentioning

confidence: 99%

Language Acquisition: The Emergence of Words from Multimodal Input

Bosch

Boves

2008

Text, Speech and Dialogue

View full text Add to dashboard Cite

Abstract. Young infants learn words by detecting patterns in the speech signal and by associating these patterns to stimuli provided by nonspeech modalities (such as vision). In this paper, we discuss a computational model that is able to detect and build word-like representations on the basis of multimodal input data. Learning of words (and wordlike entities) takes place within a communicative loop between a 'carer' and the 'learner'. Experiments carried out on three different European languages (Finnish, Swedish, and Dutch) show that a robust word representation can be learned in using approximately 50 acoustic tokens (examples) of that word. The model is inspired by the memory structure that is assumed functional for human speech processing.

show abstract

Evolving fuzzy systems for data streams: a survey

Baruah

Angelov

2011

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Evolving fuzzy systems (EFSs) can be regarded as intelligent systems based on fuzzy rule-based or neuro-fuzzy models with the ability to learn continuously and to gradually develop with the objective of enhancing their performance. Such systems learn in online mode by analyzing incoming samples, and adjusting both structure and parameters. The objective of this chapter is to present a brief overview of some early as well as recent EFSs by focusing on their architecture, design algorithms along with the merits and demerits, and various applications.

show abstract

Incremental learning with partial instance memory

Cited by 102 publications

References 25 publications

Using Adaptive Alert Classification to Reduce False Positives in Intrusion Detection

Using Adaptive Alert Classification to Reduce False Positives in Intrusion Detection

Language Acquisition: The Emergence of Words from Multimodal Input

Evolving fuzzy systems for data streams: a survey

Contact Info

Product

Resources

About