本文選題：有序回歸 + 核方法�。� 參考：《南京航空航天大學(xué)》2016年碩士論文

【摘要】：有序回歸(Ordinal Regression,OR)是模式識(shí)別中一種特殊的有監(jiān)督學(xué)習(xí)范例,其目的是基于一組給定的輸入和離散有序的輸出數(shù)據(jù)建立一個(gè)回歸器,使其能實(shí)現(xiàn)對(duì)相應(yīng)有序離散測(cè)試樣本類(lèi)標(biāo)的預(yù)測(cè)。在現(xiàn)實(shí)應(yīng)用中,此種類(lèi)標(biāo)離散而有序的場(chǎng)景廣泛存在,這些類(lèi)標(biāo)值通常是基于人們的偏好而被賦予的。此類(lèi)標(biāo)號(hào)不僅體現(xiàn)了有序標(biāo)號(hào)間的等級(jí)差異,同時(shí)也反映了具有不同類(lèi)標(biāo)樣本之間的聯(lián)系。因而不同于單純的回歸與分類(lèi)學(xué)習(xí),OR兼具分類(lèi)和回歸的雙重特性。相比于普通回歸方法,有序回歸模型融合利用了類(lèi)標(biāo)號(hào)的有序性這一先驗(yàn)信息,使得其具有比普通回歸更優(yōu)的回歸性能。近年來(lái),因OR在人臉識(shí)別、信用評(píng)級(jí)、腦機(jī)接口以及年齡估計(jì)等領(lǐng)域的廣泛應(yīng)用,其得到了研究者們?cè)絹?lái)越多的關(guān)注和研究。然而,現(xiàn)有OR方法在問(wèn)題先驗(yàn)信息的利用方面仍存在某些遺漏的不足。為此,本文針對(duì)性地在以下幾方面展開(kāi)了研究:鑒于現(xiàn)有OR方法對(duì)有序標(biāo)號(hào)間臨近關(guān)聯(lián)信息利用不足的問(wèn)題,我們?cè)O(shè)計(jì)出了一類(lèi)有序標(biāo)號(hào)間關(guān)聯(lián)度的量化表示,然后通過(guò)將該關(guān)聯(lián)度表示與經(jīng)典且有效的有序回歸方法KDLOR(Kernel Discriminant Learning for Ordinal Regression)相結(jié)合為有序?qū)W習(xí)建模,并將構(gòu)建出的方法稱(chēng)為結(jié)合類(lèi)標(biāo)簽關(guān)聯(lián)度的有序線性判別回歸學(xué)習(xí)(Linear Discriminant Learning for Ordinal Regression using Label Membership,LM-LDLOR),為應(yīng)對(duì)非線性有序?qū)W習(xí)問(wèn)題,我們還對(duì)LM-LDLOR進(jìn)行核化從而得到結(jié)合類(lèi)標(biāo)簽關(guān)聯(lián)度的有序核判別回歸學(xué)習(xí),簡(jiǎn)記為L(zhǎng)M-KDLOR。最后,在8個(gè)標(biāo)準(zhǔn)有序回歸數(shù)據(jù)集上的對(duì)比實(shí)驗(yàn)驗(yàn)證了本文所提策略對(duì)有序回歸性能提升的有效性。緩變?cè)硎腔谌祟?lèi)的視覺(jué)特性而構(gòu)建的一種學(xué)習(xí)原理,并且在模式識(shí)別中已有了較廣泛的應(yīng)用。然而,據(jù)我們所知緩變學(xué)習(xí)原理尚未與有序回歸學(xué)習(xí)相結(jié)合進(jìn)行研究。受此啟發(fā),本文通過(guò)最近鄰法對(duì)每個(gè)樣本類(lèi)構(gòu)建多個(gè)類(lèi)內(nèi)輸出有序序列計(jì)算緩變類(lèi)內(nèi)散度矩陣,同時(shí)通過(guò)有序約束保證類(lèi)標(biāo)號(hào)的有序性,然后根據(jù)線性判別準(zhǔn)則進(jìn)行映射實(shí)現(xiàn)有序?qū)W習(xí),最終形成了基于緩變?cè)淼呐袆e有序回歸方法SP-DLOR(Slowness Principle based Discriminant Learning for Ordinal Regression)。最后,在8個(gè)標(biāo)準(zhǔn)有序回歸數(shù)據(jù)集以及人臉數(shù)據(jù)集FG-NET上的對(duì)比實(shí)驗(yàn)驗(yàn)證了所提算法在回歸和分類(lèi)性能上的優(yōu)越性。
[Abstract]:Ordinal regression order is a special supervised learning paradigm in pattern recognition, which aims to build a regression based on a given set of input and discrete ordered output data. It can realize the prediction of the corresponding ordered discrete test samples. In practical applications, this kind of label is widely used in discrete and orderly situations, and these class values are usually given based on people's preferences. This kind of label not only reflects the grade difference among the ordered labels, but also reflects the relationship between the samples with different classes. Therefore, different from the simple regression and classification learning OR has the dual characteristics of classification and regression. Compared with ordinary regression methods, the integration of ordered regression model makes use of the prior information of class labeling, which makes it have better regression performance than ordinary regression. In recent years, due to the wide application of OR in face recognition, credit rating, brain-computer interface and age estimation, it has attracted more and more attention and research. However, there are still some omissions in the use of prior information in the existing OR methods. For this reason, this paper focuses on the following aspects: in view of the insufficient use of the adjacent correlation information between ordered labels by existing OR methods, we design a kind of quantitative representation of the correlation degree between ordered labels. Then, the relation degree representation is combined with the classical and effective ordered regression method KDLOR(Kernel Discriminant Learning for Ordinal Regression to model ordered learning. The method is called Linear Discriminant Learning for Ordinal Regression using Label regression Learning (LM-LDLORA) combined with class label correlation degree. In order to deal with the problem of nonlinear ordered learning, We also nucleate LM-LDLOR and obtain an ordered kernel discriminant regression learning combined with class label correlation, which is abbreviated as LM-KDLOR. Finally, a comparative experiment on eight standard ordered regression datasets verifies the effectiveness of the proposed strategy in improving the performance of ordered regression. The principle of slow change is a learning principle based on human visual characteristics, and has been widely used in pattern recognition. However, as far as we know, the principle of slow learning has not been combined with ordered regression learning. Inspired by this, this paper constructs multiple in-class output order sequences for each sample class by nearest neighbor method to calculate the slowly varying intra-class dispersion matrix, and guarantees the ordering of class labels through ordered constraints. Then, according to the linear discriminant criterion, the ordered learning is realized by mapping, and finally a discriminant ordered regression method, SP-DLOR(Slowness Principle based Discriminant Learning for Ordinal Regression, is formed based on the principle of slow variation. Finally, the comparison experiments on eight standard ordered regression data sets and face dataset FG-NET show the superiority of the proposed algorithm in regression and classification performance.
【學(xué)位授予單位】：南京航空航天大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類(lèi)號(hào)】：TP391.4

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 賈洪杰;丁世飛;史忠植;;求解大規(guī)模譜聚類(lèi)的近似加權(quán)核k-means算法[J];軟件學(xué)報(bào);2015年11期

2 周航星;陳松燦;;有序判別典型相關(guān)分析[J];軟件學(xué)報(bào);2014年09期

3 余海r，

本文編號(hào)：1816701