Credal Fusion of Classifications for Noisy and Uncertain Data

Karem, Fatma; Dhibi, Mounir; Atkinson, Martin; Bouhlel, Med Salim

doi:10.11591/ijece.v7i2.pp1071-1087

Cited by 4 publications

(6 citation statements)

References 16 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results concludes that adding recent block of data instances to provided standard training dataset is a wise solution and improves the performance of ensemble very well. which are responsible behind the change in accuracy.Noisy, unbalance data and missing values of attributes were found to be the root cause behind the drift [23]. Our proposed system shows improved performance while detecting both kinds of drifts efficiently.However our work focuses on offline streaming data.…”

Section: Results and Analysismentioning

confidence: 91%

Concept Drift Identification using Classifier Ensemble Approach

Deshpande¹,

Rao²

2018

IJECE

View full text Add to dashboard Cite

Abstract:-In Internetworking system, the huge amount of data is scattered, generated and processed over the network. The data mining techniques are used to discover the unknown pattern from the underlying data. A traditional classification model is used to classify the data based on past labelled data. However in many current applications, data is increasing in size with fluctuating patterns. Due to this new feature may arrive in the data. It is present in many applications like sensornetwork, banking and telecommunication systems, financial domain, Electricity usage and prices based on its demand and supplyetc .Thus change in data distribution reduces the accuracy of classifying the data. It may discover some patterns as frequent while other patterns tend to disappear and wrongly classify. To mine such data distribution, traditionalclassification techniques may not be suitable as the distribution generating the items can change over time so data from the past may become irrelevant or even false for the current prediction. For handlingsuch varying pattern of data, concept drift mining approach is used to improve the accuracy of classification techniques. In this paper we have proposed ensemble approach for improving the accuracy of classifier. The ensemble classifier is applied on 3 different data sets. We investigated different features for the different chunk of data which is further given to ensemble classifier. We observed the proposed approach improves the accuracy of classifier for different chunks of data.

show abstract

Section: Results and Analysismentioning

confidence: 91%

Concept Drift Identification using Classifier Ensemble Approach

Deshpande¹,

Rao²

2018

IJECE

View full text Add to dashboard Cite

show abstract

“…Belief Function results: Also for this method a 5000 users' preferences are generated toward 10, 30, and 100 [30,29] = 1 so the two objects are indifferent. And for each pair the adjacency matrix is checked to get the information about preferences for that pair.…”

Section: Resultsmentioning

confidence: 99%

“…The mass function represents one of many functions used in this step, among them plausibility function, belief function, communality function etc. can be cited [27], [29].…”

Section: Preferences Fusion Based On the Dempster Shafermentioning

confidence: 99%

See 1 more Smart Citation

Majority Vote method for preferences detection: Application for Social Networks

2019

IJATCSE

View full text Add to dashboard Cite

Most of recommender systems rely on the users' preferences to recommend items. With the increase of data and the number of users on the internet, the task of recommender system becomes more and more sophisticated. The fusion of users' preferences is a solution to facilitate that task. The resolution of this issue, helps in decreasing the run time of a recommender system. This research focuses on the issue of preference fusion in the context of social network. The aim of this research is to determine the collective preferences of users belonging to a given community based on their individual preferences. In particular, researchers are interested here in the representation of individual preferences, taking into account the uncertainty and imprecision phenomenon. In this research two methods are proposed: majority vote method and the Dempster Shafer theory of evidence. Experiments on generated datasets would highlight the interest of the proposed solution.

show abstract

“…The fusion of unimodal multi-feature and one-dimensional hidden markov models for low-resolution face recognition as in [9]. Also it is used in many applications like credal fusion of classifications for noisy and uncertain data and texture fusion for batik motif retrieval system as on [10], [11].…”

Section: Introductionmentioning

confidence: 99%

An Image Fusion Method Based On Framelet Transform

Dhanji¹,

Singh²

2018

IJEECS

View full text Add to dashboard Cite

<p>Image fusion is the process of combining relevant information from two or more images into a single image. The resulting image contains more information as compared to individual images. In this system, we are proposing a new image fusion method by using a technique called framelet transform. This method consists of two phases; First is the frame separation where the pre-processing and Selection of frames for both the images occurs and the second phase is the Frame fusion that is done by inverse framelet transform method is done from which a fused output images will be formed that will be containing both the information of the input images (CT & MRI images).</p>

show abstract

Credal Fusion of Classifications for Noisy and Uncertain Data

Cited by 4 publications

References 16 publications

Concept Drift Identification using Classifier Ensemble Approach

Concept Drift Identification using Classifier Ensemble Approach

Majority Vote method for preferences detection: Application for Social Networks

An Image Fusion Method Based On Framelet Transform

Contact Info

Product

Resources

About