Pattern classification with missing data: a review

García-Laencina, Pedro J.; Sancho‐Gómez, José‐Luis; Figueiras-Vidal, Anı́bal R.

doi:10.1007/s00521-009-0295-6

Cited by 646 publications

(434 citation statements)

References 53 publications

Supporting

Mentioning

388

Contrasting

Unclassified

Order By: Relevance

“…Sınıflandırma ve tahmin etmede karar ağaçlarının kullanılması, eğitim verisinden karar ağacı modelinin oluşturulması, bu modelin, test verisi kullanılarak uygun sınama ölçütleri aracılığıyla değerlendirilmesi ve ilgili modelin gelecekteki değerleri tahmin edilmesinde kullanılması şeklinde işlemektedir [23][24].…”

Section: Method)unclassified

Şirket İflaslarının Tahminlenmesinde Karar Ağacı Algoritmalarının Karşılaştırmalı Başarım Analizi

Onan¹

2015

BTD

View full text Add to dashboard Cite

Özet-Bu çalışmada, önemli bir ekonomik problem olan, şirket iflaslarının tahmin edilmesi ele alınmıştır. Bunun için iki yüz kırk farklı şirkete ilişkin finansal özellikleri içeren bir veri seti kullanılmıştır. Ele alınan veri seti, sınıflandırma ve tahmin etmede kullanılan önemli yöntemlerden biri olan karar ağacı yöntemine ilişkin yedi farklı algoritma uygulanarak, doğru sınıflandırma yüzdesi, ortalama mutlak hata, ortalama karesel hatanın karekökü, kesinlik, geri çağırma, F-ölçütü gibi ölçütler bakımından değerlendirilmiştir. Deneysel sonuçlar incelendiğinde, karar ağacı algoritmalarının şirket iflaslarının tahmin edilmesi için uygun bir yöntem olduğu ve kısmen başarılı doğru sınıflandırma yüzdesi elde ettiği gözlemlenmiştir.Anahtar Kelimeler-tahmin etme, karar ağacı algoritmaları, sınıflandırma, şirket iflasları Comparative Performance Analysis of Decision Tree Algorithms in the Corporate Bankruptcy PredictionAbstract-In this study, corporate bankruptcy prediction, a crucial economic problem is tackled. To do this, a data set of 240 distinct companies with financial features is used. This data set is applied to one of the most important classification and forecasting methods, i.e. decision tree method. Seven different decision tree algorithms are evaluated in terms of accuracy percentage, mean absolute error, root mean squared error, precision, recall, F-measure. According to experimental results, decision tree algorithms are appropriate methods for corporate bankruptcy prediction with relatively successful accuracy rates.

show abstract

Section: Method)unclassified

Şirket İflaslarının Tahminlenmesinde Karar Ağacı Algoritmalarının Karşılaştırmalı Başarım Analizi

Onan¹

2015

BTD

View full text Add to dashboard Cite

show abstract

“…The suggested method is based on the well-known k nearest-neighbors imputation procedure [11,25,26] that has been successfully demonstrated for missing data imputation in forestry remote sensing, where missing forest-related characteristics were imputed using the Landsat thematic mapper TMand ETM+ satellite image data [27][28][29][30][31] or airborne light detection and ranging LiDARimage data [32][33][34].…”

Section: Sensor-to-sensor Prediction (Sentos) Methodsmentioning

confidence: 99%

“…This procedure is often called imputation [9], in general, or image inpainting [8,10] in the field of image processing. Missing data can be imputed with statistical methods [9], machine-learning methods or model-based methods [11]. The missing data problem can be handled by two approaches: "within sensor" and "between sensors".…”

Section: Approaches To Handling the Missing Data Problemmentioning

confidence: 99%

Using Visible Spectral Information to Predict Long-Wave Infrared Spectral Emissivity: A Case Study over the Sokolov Area of the Czech Republic with an Airborne Hyperspectral Scanner Sensor

et al. 2013

View full text Add to dashboard Cite

Remote-sensing platforms are often comprised of a cluster of different spectral range detectors or sensors to benefit from the spectral identification capabilities of each range. Missing data from these platforms, caused by problematic weather conditions, such as clouds, sensor failure, low temporal coverage or a narrow field of view (FOV), is one of the problems preventing proper monitoring of the Earth. One of the possible solutions is predicting a detector or sensor's missing data using another detector/sensor. In this paper, we propose a new method of predicting spectral emissivity in the long-wave infrared (LWIR) spectral region using the visible (VIS) spectral region. The proposed method is suitable for two main scenarios of missing data: sensor malfunctions and narrow FOV. We demonstrate the usefulness and limitations of this prediction scheme using the airborne hyperspectral scanner (AHS) sensor, which consists of both VIS and LWIR spectral regions, in a case study over the Sokolov area, Czech Republic.

show abstract

“…The effectiveness of PCC method has been tested through four experiments with artificial and real data sets. Pedro J.Gracia-Laencina, Jose-Luis Sancho-Gomez [2] proposes Pattern classification with achievement utilized as a part of a few problem areas, as biometric acknowledgment, record classification or analysis. Missing data is a standard inconvenience that example acknowledgment systems are constrained to adjust once determination genuine assignments classification.…”

Section: IImentioning

confidence: 99%

Implementation of Hierarchical Clustering for Improved Classification of Incomplete Pattern

Thakre¹

2017

IJRASET

View full text Add to dashboard Cite

Most of the time values are missing in database, which should be dealt with. Missing qualities are happened in light of the fact that, the information section individual did not know the correct esteem or disappointment of sensors or leave the space purge. The arrangement of missing esteemed deficient example is a testing errand in machine learning approach. Fragmented information is not appropriate for classification handle. At the point when inadequate examples are arranged utilizing prototype values, the last class for similar examples may have different outcomes that are variable yields. We cannot characterize particular class for particular examples. The framework creates a wrong outcome which additionally brings about differing impacts. So to manage such sort of inadequate information, framework executes prototype-based credal classification (PCC) technique. The PCC technique is fused with Hierarchical bunching and Evidential thinking strategy to give exact, time and memory productive results. This technique prepares the specimens and recognizes the class prototype. This will be helpful for identifying the missing qualities. At that point in the wake of getting every single missing worth, credal strategy is use for classification. The trial comes about demonstrate that the improved form of PCC performs better as far as time and memory effectiveness. Keywords: Belief functions, hierarchical clustering, credal classification, evidential reasoning, missing data. I.INTRODUCTION Information mining can be considered as a strategy to discover appropriate data from extensive datasets and distinguishing designs. Such examples are further helpful for classification handle. The fundamental usefulness of the information mining procedure is to discover helpful data inside dataset and change over it into an educated organization for some time later. In a large portion of the classification issue, some quality fields of the protest are vacant. There are different explanation for the void characteristics including disappointment of sensors, mistaken qualities field by client, at some point didn't get the significance of field so client leave that field exhaust and so forth. There is a need to discover the proficient technique to characterize the protest which has missing trait values. Different classification techniques are accessible in writing to manage the classification of inadequate examples. Some system evacuates the missing esteemed examples and just uses finish designs for the classification procedure. Be that as it may, at some point deficient examples contain critical data accordingly this strategy is not a legitimate arrangement. Additionally this strategy is material just when deficient information is under 5% of entire information. Overlooking the fragmented information may diminish the quality and execution of classification calculation. Next technique is just to fill the missing qualities however it is additionally tedious process. This paper is based on the classification of fragmented patterns. ...

show abstract

Pattern classification with missing data: a review

Cited by 646 publications

References 53 publications

Şirket İflaslarının Tahminlenmesinde Karar Ağacı Algoritmalarının Karşılaştırmalı Başarım Analizi

Şirket İflaslarının Tahminlenmesinde Karar Ağacı Algoritmalarının Karşılaştırmalı Başarım Analizi

Using Visible Spectral Information to Predict Long-Wave Infrared Spectral Emissivity: A Case Study over the Sokolov Area of the Czech Republic with an Airborne Hyperspectral Scanner Sensor

Implementation of Hierarchical Clustering for Improved Classification of Incomplete Pattern

Contact Info

Product

Resources

About