Classification of internet users using discriminant analysis and neural networks

Nogueira, António; Oliveira, M. Rosário; Salvador, Paulo; Valadas, Rui; Pacheco, Antònio

doi:10.1109/ngi.2005.1431686

Cited by 19 publications

(12 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several classification units can be implemented: Artificial Neural Networks (ANN), Fisher Linear Discriminant Analysis (FLD), Spectral Angle Mapper (SAM) and K-Nearest Neighbors (KNN). ANNs were initially employed because of their ability to handle non-linearity, their parallel processing of information and their quick adaptability to system dynamics (Nogueira, 2005). ANNs turned out to be very useful in this foreign object detection application and high classification accuracies (Garcia-Allende, 2007) were achieved, suggesting that maybe simpler algorithms could be employed instead.…”

Section: His Classification Alternatives For Quality Controlmentioning

confidence: 99%

Hyperspectral Imaging for Raw Material Sorting and Processed Product Quality Control

García-Allende¹,

Conde²,

López‐Higuera³

2011

Applications and Experiences of Quality Control

View full text Add to dashboard Cite

Section: His Classification Alternatives For Quality Controlmentioning

confidence: 99%

Hyperspectral Imaging for Raw Material Sorting and Processed Product Quality Control

García-Allende¹,

Conde²,

López‐Higuera³

2011

Applications and Experiences of Quality Control

View full text Add to dashboard Cite

“…19 Since the large number of papers published in attempts to establish the relative superiority of any of these algorithms have failed, any of them could be employed for the identification task. Artificial Neural Networks (ANN), 20 because of their ability to handle non-linearity, their parallel processing of information and their quick adaptability to system dynamics, 21 were initially employed in material discrimination. ANNs proved, with the appropriate training, very accurate in spurious material detection.…”

Section: Hyperspectral Image Discriminationmentioning

confidence: 99%

Hyperspectral imaging for diagnosis and quality control in agri-food and industrial sectors

García-Allende

Conde

Mirapeix

et al. 2010

Optical Sensing and Detection

View full text Add to dashboard Cite

Optical spectroscopy has been utilized in various fields of science, industry and medicine, since each substance is discernible from all others by its spectral properties. However, optical spectroscopy traditionally generates information on the bulk properties of the whole sample, and mainly in the agri-food industry some product properties result from the heterogeneity in its composition. This monitoring is considerably more challenging and can be successfully achieved by the so-called hyperspectral imaging technology, which allows the simultaneous determination of the optical spectrum and the spatial location of an object in a surface. In addition, it is a nonintrusive and non-contact technique which gives rise to a great potential for industrial applications and it does not require any particular preparation of the samples, which is a primary concern in food monitoring. This work illustrates an overview of approaches based on this technology to address different problems in agri-food and industrial sectors. The hyperspectral system was originally designed and tested for raw material on-line discrimination, which is a key factor in the input stages of many industrial sectors. The combination of the acquisition of the spectral information across transversal lines while materials are being transported on a conveyor belt, and appropriate image analyses have been successfully validated in the tobacco industry. Lastly, the use of imaging spectroscopy applied to online welding quality monitoring is discussed and compared with traditional spectroscopic approaches in this regard.

show abstract

“…Neural Networks (NNs) have been successfully used in several different fields due to their advantageous properties like parallel processing of information, capacity to handle non-linearity and quick adaptability to system dynamics: pattern recognition in the presence of noise and non-linearity [28]- [30], classification of Internet users [31], [32], intrusion detection [33], [34], among other applications. Neural Networks can also be used to identify Internet applications, and particularly P2P applications, based on their characteristic traffic patterns.…”

Section: Identification Framework Based On Neural Networkmentioning

confidence: 99%

Towards the On-line Identification of Peer-to-peer Flow Patterns

Nogueira¹,

Salvador²,

Couto³

et al. 2009

JNW

View full text Add to dashboard Cite

The number and variety of IP applications have hugely increased in the last few years. Among them, peer-to-peer (P2P) file-sharing applications have become very popular: more users are continuously joining such systems and more objects are being made available, seducing even more users to join. An accurate mapping of traffic to applications is important for a wide range of network management tasks. Besides, traditional mapping approaches have become increasingly inaccurate because many applications use non-default or ephemeral port numbers, use well-known port numbers associated with other applications, change application signatures or use traffic encryption. This paper proposes a framework to identify Internet applications that can be mainly used in situations where existing identification frameworks are not efficient or can not be used at all. The core block of the identification tool is based on neural networks and is able to identify different flow patterns generated by various Internet applications. Neural network based identification relies on a previous identification of the different IP applications that can be obtained offline using any reliable method. In this way, the paper also presents a module to process IP traffic flows and identify the underlying applications using payload analysis techniques. The identification results obtained from this tool are used in the training phase of the neural network identification framework. The accuracy of the identification framework was evaluated by performing a set of intensive tests and the results obtained show that, when conveniently trained, neural networks constitute a valuable tool to identify Internet applications while being, at the same time, immune to the most important disadvantages presented by other identification methods.

show abstract

Classification of internet users using discriminant analysis and neural networks

Cited by 19 publications

References 13 publications

Hyperspectral Imaging for Raw Material Sorting and Processed Product Quality Control

Hyperspectral Imaging for Raw Material Sorting and Processed Product Quality Control

Hyperspectral imaging for diagnosis and quality control in agri-food and industrial sectors

Towards the On-line Identification of Peer-to-peer Flow Patterns

Contact Info

Product

Resources

About