35 Hits in 7.6 sec

The Dynamics of AdaBoost Weights Tells You What's Hard to Classify [article]

Bruno Caprile, Cesare Furlanello & Stefano Merler
2002 arXiv   pre-print
The dynamical evolution of weights in the Adaboost algorithm contains useful information about the role that the associated data points play in the built of the Adaboost model.  ...  In particular, the dynamics induces a bipartition of the data set into two (easy/hard) classes.  ...  The Dynamics of Weights In the present section, the dynamics that the AdaBoost algorithm sets over the weights is singled out for study.  ... 
arXiv:cs/0201014v1 fatcat:vs5hfi3gwrgplgpnt3shl7yof4

Learning Microarray Gene Expression Data by Hybrid Discriminant Analysis

Yijuan Lu, Qi Tian, Maribel Sanchez, Jennifer Neary, Feng Liu, Yufeng Wang
2007 IEEE Multimedia  
We conducted a set of 80 microarray experiments to test this technique as well as a boosted hybrid analysis technique. 22 1070-986X/07/$25.00  ...  The intention is to increase the weights of the incorrectly classified examples and decrease the weights of the correctly classified examples.  ...  But it's hard to tell which parameter is best. Searching the whole parametric space will result in extra computational complexity.  ... 
doi:10.1109/mmul.2007.78 fatcat:2onxpob4c5c65f23fcnyulwhxq

A survey of tagging techniques for music, speech and environmental sound

Shufei Duan, Jinglan Zhang, Paul Roe, Michael Towsey
2012 Artificial Intelligence Review  
This survey aims to provide an overview about the state-of-the-art development in these areas. We discuss about the meaning of tagging in different sound areas at the beginning of the journey.  ...  Some examples of sound tagging applications are introduced in order to illustrate the significance of this research.  ...  By counting the number of different types of tags for the same music clip, a weight score will be added to the tag and then put into classifiers.  ... 
doi:10.1007/s10462-012-9362-y fatcat:7g72qonsvfer5awzsawbwxmiz4

Face Recognition System [article]

Yang Li, Sangwhan Cha
2019 arXiv   pre-print
For different applications (images, voice, text), you need to use different network models to achieve better results.  ...  Deep learning can be classified as a neural network from the general category, but there are many changes in the concrete realization.  ...  Applications of face recognition in the real world is hard to see.  ... 
arXiv:1901.02452v1 fatcat:7orsmcfomzbdjmqnuyhi2mgj64

A lifelong spam emails classification model

Rami Mustafa A. Mohammad
2020 Applied Computing and Informatics  
prone to the so-called "Concept Drift".  ...  For that reason, even though traditional classification methods possess sound classification results, they were ineffective for lifelong classification of spam emails duo to the fact that they might be  ...  In general, growing datasets might cause a model working in dynamic situations, that might result in the what's normally known as "Concept Drift" issue [18, 19] .  ... 
doi:10.1016/j.aci.2020.01.002 fatcat:biwpczx3zveshf2ug4dmjz7iiy

ijair-volume-6-issue-1-vii-january-march-2019 -HINDUSTAN BOOK.pdf

V. Thamilarasi
Feed forward neural network where each node outputs an activation function applied over the weighted sum of its inputs Weighted Associative Classifier (WAC) Weighted Associative Classifier is a new concept  ...  index filter is used.  Pixel based detecting the white nodule likeliness map done using Laplace Gaussian blob detection.  By using Adaboost classifier classifies and ranking according to the combination  ...  like IPv6 and 5G to achieve the dynamic mash up of IoT topology.  ... 
doi:10.6084/m9.figshare.20217722.v1 fatcat:l74ihuqhcvdtjomod3zdwzfniu

Intelligence, physics and information – the tradeoff between accuracy and simplicity in machine learning [article]

Tailin Wu
2020 arXiv   pre-print
How can we enable machines to make sense of the world, and become better at learning?  ...  I believe that building on the work of my thesis we will be one step closer to enable more intelligent machines that can make sense of the world.  ...  We use noisy label to mimic realistic settings where the data may be noisy and also to have controllable difficulty for different classes.  ... 
arXiv:2001.03780v2 fatcat:piduzlhoafcjhhsgthulbbhtke

Exploring and Exploiting Diversity for Image Segmentation [article]

Payman Yadollahpour
2017 arXiv   pre-print
The formulation is agnostic to the underlying segmentation model (e.g. CRF, CNN, etc.) and optimization algorithm, which makes it applicable to a wide range of models and inference methods.  ...  Due to limitations in optimization, the model cannot be arbitrarily complex.  ...  In each stage the classifier is a combination of a set of weak classifiers that has been trained using AdaBoost [32] .  ... 
arXiv:1709.01625v1 fatcat:hjedzbi37be5hk4pai4hc3oany

Comparative Analysis of Different Machine Learning Classifiers for the Prediction of Chronic Diseases [chapter]

Rajesh Singh, Anita Gehlot, Dharam Buddhi
2022 Comparative Analysis of Different Machine Learning Classifiers for the Prediction of Chronic Diseases  
This paper forms the basis of understanding the difficulty of the domain and the amount of efficiency achieved by the various methods recently.  ...  A comparative study of different machine learning classifiers for chronic disease prediction viz Heart Disease & Diabetes Disease is done in this paper.  ...  The performance of existing and proposed algorithms is analysed with regard to several metrics.  ... 
doi:10.13052/rp-9788770227667 fatcat:da47mjbbyzfwnbpde7rgbrlppe

Multi-Modal Music Information Retrieval: Augmenting Audio-Analysis with Visual Computing for Improved Music Video Analysis [article]

Alexander Schindler
2020 arXiv   pre-print
This thesis focuses on the information provided by the visual layer of music videos and how it can be harnessed to augment and improve tasks of the MIR research domain.  ...  The main hypothesis of this work is based on the observation that certain expressive categories such as genre or theme can be recognized on the basis of the visual content alone, without the sound being  ...  You have to be given what's needed by nature, and what's needed is to bring something new."  ... 
arXiv:2002.00251v1 fatcat:6cz6rivc3fbg7fahdsnokxfrk4

Text-based Sentiment Analysis and Music Emotion Recognition [article]

Erion Çano
2018 arXiv   pre-print
Second, there are various uncertainties regarding the use of word embedding vectors: should they be generated from the same data set that is used to train the model or it is better to source them from  ...  This thesis addresses the above problems to provide methodological and practical insights for utilizing neural networks on sentiment analysis of texts and achieving state of the art results.  ...  This high number of parameters makes it hard to find the optimal performance setup.  ... 
arXiv:1810.03031v1 fatcat:4vj4euwtxbghbjdev2gutcgjny

If I Hear You Correctly: Building and Evaluating Interview Chatbots with Active Listening Skills [article]

Ziang Xiao, Michelle X. Zhou, Wenxi Chen, Huahai Yang, Changyan Chi
2020 pre-print
To evaluate the effectiveness of our prototype, we compared the performance of interview chatbots with or without active listening skills on four common interview topics in a live evaluation with 206 users  ...  As the first step, we are investigating the feasibility and effectiveness of using publicly available, practical AI technologies to build effective interview chatbots.  ...  ACKNOWLEDGMENT This work is in part supported by the Air Force Office of Basic Research under FA9550-15-C-0032.  ... 
doi:10.1145/3313831.3376131 arXiv:2002.01862v1 fatcat:jn6aqz3pafg4zeucufuhqpk3bi

A face tracking system for dynamic event recognition

Nicholas Michael
We then move onto the focus of our work, which is the application of the face tracker to two interesting recognition problems.  ...  Therefore, we present our novel framework for robust tracking and analysis of facial expressions and head gestures, by means of a dynamic feature descriptor, a 3D face model and temporal models, with an  ...  , yielding normalized synchrony features, which you used to train classifiers to detect deceptive and truth-telling subjects.  ... 
doi:10.7282/t3mc8z1j fatcat:mxsl2falsfhnhcgcv77ag6af6i

Is Urban Space Able to Affect Community Participation? a Research Proposal between Urban Sociology and Environmental Psychology

Luca Bottini
2015 European Journal of Social Sciences Education and Research  
This different point of view allow us to deepen the relation within spaces and people, measuring several aspect of the impact of space in individual's life, such as place attachment, urban quality perceived  ...  This paper will deal with the study of the relation between the urban spaces and social phenomena.  ...  Therefore, it is the purpose of this paper to investigate a different aspect, the involvement of the four (4) Greek EFL School Advisors who acted as Coordinators of the four (4) online CoPs under research  ... 
doi:10.26417/ejser.v4i1.p81-86 fatcat:hi7ewbbnb5hlpgq3qtmdzmpgsy

Learning in the Real World: Constraints on Cost, Space, and Privacy

Matt J. Kusner
a weighted classifier with cost-weighted 0regularization.  ...  The cascade consists of stages of Adaboost classifiers that either 'reject' or 'pass on' inputs for further classification.  ...  Recursively solving these problems within a classifier tree allows us to refine the features we select to rank each web-page, instead of simply selecting the same set of features for all inputs.  ... 
doi:10.7936/k7xs5tsp fatcat:vrctfharxbcblayuo24gnq6rai
« Previous Showing results 1 — 15 out of 35 results