tag:blogger.com,1999:blog-37324607.post2511612350739234558..comments2019-11-08T02:00:08.169-05:00Comments on Data Mining in MATLAB: Principal Components AnalysisWill Dwinnellhttp://www.blogger.com/profile/03379859054257561952noreply@blogger.comBlogger57125tag:blogger.com,1999:blog-37324607.post-11603403467843632222019-03-10T07:29:27.672-04:002019-03-10T07:29:27.672-04:00Hi, Will!
Nice explanation! Thanks!
Could you al...Hi, Will!<br /><br />Nice explanation! Thanks!<br /><br />Could you also give an explanation for the case when the dimensionality (d) is greater than the number of samples (s), and how to extract more PCs than s?<br /><br />For example:<br />Given:<br />s = 100<br />d = 3000<br /><br />Problem:<br />Let's say reduce d to 1500.<br /><br />Thanks in advance<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-60093899856890760022017-09-21T06:08:11.850-04:002017-09-21T06:08:11.850-04:00This was very resourceful. Thanks! This was very resourceful. Thanks! Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-3002371679673840642016-10-22T23:35:19.819-04:002016-10-22T23:35:19.819-04:00Man! It helps a lot! Thank you very much!Man! It helps a lot! Thank you very much!Bj Wuhttps://www.blogger.com/profile/16198338691567902087noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-85631901704663797572016-01-21T12:11:46.920-05:002016-01-21T12:11:46.920-05:00Hey thank u for an example.I tried it with and wit...Hey thank u for an example.I tried it with and without using princomp function and my solution without using princomp matches with your answer but i am getting different answer when i use princomp as follows:<br />coeff =<br /><br /> 0.9686 0.1808 -0.1709<br /> 0.2019 -0.1698 0.9646<br /> -0.1454 0.9687 0.2010<br /><br /><br />score =<br /><br /> -0.1721 -0.0108 0.0272<br /> 2.5399 -0.1270 0.0612<br /> 0.0216 0.0253 -0.0070<br /> 2.0831 0.0284 -0.0232<br /> -0.1721 -0.0108 0.0272<br /> -0.1721 -0.0108 0.0272<br /> -1.7388 -0.5398 -0.0491<br /> -1.8260 0.0414 0.0715<br /> -3.1127 0.1830 -0.0490<br /> -2.2829 0.1968 -0.0129<br /> 3.7224 0.0730 -0.0473<br /> 1.2388 0.1115 -0.0392<br /> -0.1721 -0.0108 0.0272<br /> 0.0216 0.0253 -0.0070<br /> 0.0216 0.0253 -0.0070<br /><br /><br />latent =<br /><br /> 3.3971<br /> 0.0287<br /> 0.0015<br /><br />Please help me out!<br />Thank youSaranya Bharathihttps://www.blogger.com/profile/01885350040629273716noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-7300110584732017012015-12-06T11:50:45.050-05:002015-12-06T11:50:45.050-05:00I have been going through number of websites and t...I have been going through number of websites and textbooks to learn how to utilize PCA and somewhat confused because of so many technical terms and formulas flying around. Your explanation just settled everything in place in my brain. Thank you!Akirahttps://www.blogger.com/profile/05378336842888608959noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-1901612948861883262015-11-19T02:09:09.755-05:002015-11-19T02:09:09.755-05:00hai,will can we find the attribute names which has...hai,will can we find the attribute names which has been reduced as the result...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-34461510598881412552015-09-08T02:45:04.726-04:002015-09-08T02:45:04.726-04:00For a multi variables data sets is it possible to ...For a multi variables data sets is it possible to tell which variable have most influence on 1st Principal component. Another question is that what is the significance of the negative sign in the V matrix component?Debasis Janahttps://www.blogger.com/profile/08817686439248454118noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-27377559861021155262015-06-08T16:06:38.997-04:002015-06-08T16:06:38.997-04:00Nice article.
After applying PCA , if I want to t...Nice article.<br /><br />After applying PCA , if I want to take only first component and throw out other 2 components and want to reproduce the original data , then my reproduced data still having three variable. Can you please tell me how this procedure affects my data.<br /><br />Or if I want to predict some Y variable based on given data , then how can we use PCA for that ?<br /><br />Thank you.munjalhttps://www.blogger.com/profile/18283210312291269744noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-9917049391244139272015-04-09T18:26:21.977-04:002015-04-09T18:26:21.977-04:00Hi Will,
Nice post to explain PCA. I wonder if yo...Hi Will, <br />Nice post to explain PCA. I wonder if you can help my simple problem. I wish to do a GPR with input from PCA of my data, and I learned that the right way to do the CV is by doing PCA on the training set, then use the training regression coefficients to map the test set to their PCs. The following is my attempt in matlab:<br /><br />% Calculate xscores for training set <br />[coef, score, latent, explained, mu] = pca(data_train(:,2:end));<br />xscore_train=score; <br />ytrain=zdata_train(:,1);<br /><br />% Calculate xscores for test set=standardized newX * coef<br />xscore_test=zdata_test(:,2:end)*coef; ytest=zdata_test(:,1);<br /><br />Do I compute the xscore_test correctly? Also, the scores returned are not standardized correct?<br /><br />Thanks a bunch:)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-39821520294783003212014-12-25T12:35:21.752-05:002014-12-25T12:35:21.752-05:00i was searching for 3 months how to use PCA for fe...i was searching for 3 months how to use PCA for feature reduction. Thank you for the content.. sdpAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-44438904505074083172014-12-16T17:00:49.529-05:002014-12-16T17:00:49.529-05:00Best explanation of PCA on the net! I've been ...Best explanation of PCA on the net! I've been googling it all day.Christopher Harrishttps://www.blogger.com/profile/16726301179207891280noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-76968966243085424632014-10-07T21:18:41.587-04:002014-10-07T21:18:41.587-04:00Does the dividing of the data by the standard devi...Does the dividing of the data by the standard deviation mess it up? Will it lower the energy of the high variance dimensions and increase the energy of the low variance dimensionsaaaaaaghhhhhhhhhhhhhhhhhhttps://www.blogger.com/profile/18444828982623599842noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-39197210606253029052014-08-27T00:55:29.154-04:002014-08-27T00:55:29.154-04:00Thanks for sharing this valuable information. I th...Thanks for sharing this valuable information. I think this PCA method is also helpful for data mining process Containing data compression.Anisha K Shttp://www.offshoreindiadataentry.com/data-mining.phpnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-34739939968537391242014-07-14T06:44:27.113-04:002014-07-14T06:44:27.113-04:00Lets assume you've 3 features of an image inst...Lets assume you've 3 features of an image instead of 3 measurements taken of a particular subject. I have 10 images. My training dataset will be 10 x 3; If I use matlab buildin function princomp and get COEFF SCORE LATENT? which one should I use; score also gives me 3 col. Do I need to use first col. only. How to use this number for better interpretation of my results? how to give input to the classifierAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-30824837171225395412014-05-21T21:35:24.959-04:002014-05-21T21:35:24.959-04:00Answering post above. B'B is 14 times cov(B) b...Answering post above. B'B is 14 times cov(B) because the covariance estimator is 1/n-1* B'B when E(B)=0 ie series are zero mean. since n=15, n-1 =14. mystery solved. very interesting. covariance esitmator link http://www.encyclopediaofmath.org/index.php/Covariance_matrixAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-22335025553872672832014-05-21T18:12:31.159-04:002014-05-21T18:12:31.159-04:00If you want to get the original data from B , you ...If you want to get the original data from B , you do not need COEFF at all. COEFF*COEFF'=identity matrix or 1. I think you should point out that V or COEFF are the eigenvectors of Cov(B). It took me a while to figure out. In this case cov(B) is same as corr(B), as they are zscores. However, B*B' is 14*cov(B). why the 14? it is puzzling me. I was thinking that will give the covariance matrix. Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-8498742932438997972014-05-21T04:07:15.899-04:002014-05-21T04:07:15.899-04:00After applying PCA on original and modified data s...After applying PCA on original and modified data sets, I want to <b>analyse the new data sets for accuracy and information loss & entropy</b>. Any functions or specific codes for the same?Nethravati Ma'mnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-37736616610644360652014-05-16T07:24:27.106-04:002014-05-16T07:24:27.106-04:00Nice post; the examples make it all very clear. I ...Nice post; the examples make it all very clear. I recently wrote an article about what PCA actually means. This might be helpful for some, to get a more intuitive understanding: http://www.visiondummy.com/2014/05/feature-extraction-using-pca/Vincent Spruythttps://www.blogger.com/profile/11514746260073858823noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-11922345686015436612014-05-13T23:32:05.650-04:002014-05-13T23:32:05.650-04:00hello please tell me how can use
pca for privacy ...hello please tell me how can use <br />pca for privacy preserving in data mining thank u.N P NETHRAVATHIhttps://www.blogger.com/profile/12718566638338069920noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-26907872788058811962014-04-03T18:37:38.144-04:002014-04-03T18:37:38.144-04:00Very good and clear, after reading this can unders...Very good and clear, after reading this can understand PCA.<br />How to use PCA to unbalanced data ? Lifehttps://www.blogger.com/profile/14356444320625183332noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-29365723703143496872014-03-20T06:03:17.909-04:002014-03-20T06:03:17.909-04:00Super blog post. It helped a lot.
However, I am ...Super blog post. It helped a lot. <br /><br />However, I am trying to get the same results out of MATLAB that a colleague is getting out of SPSS. <br /><br />I tried using applying eig to both the covariance matrix (cov) and the corelation matrix (corrcoef) but it results in the same set of eigenvectors. <br /><br />I know SPSS has the option to rotatefactors (as does MATLAB of course). But I am talking about the unrotated factors.<br /><br />Hope you can help<br /><br />Steve Westland<br />University of Leeds<br /><br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-80938131647659772482013-09-10T03:48:36.548-04:002013-09-10T03:48:36.548-04:00Thank you very much for this explanation. I found ...Thank you very much for this explanation. I found your page because I've been struggling for a while to understand why the coefficients given by matlab are systematically smaller than those given by Statistica. Today, I've realized that the norms of the factors (columns of COEFF)which should be equal to the factors' variances are instead normalized to one. Do you know why, and more importantly how I can prevent this normalization in order to one to stick to Statistica's results ?<br />Thanks in advance !<br />Isabelle Shttps://www.blogger.com/profile/05661851598386894447noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-31591525030820933662013-08-06T07:46:24.557-04:002013-08-06T07:46:24.557-04:00The PC's are orthogonal (V*V.'), but I may...The PC's are orthogonal (V*V.'), but I may say that the transformed data is not always uncorrelated. In case, in the original data, there is one variable that is a linear combination of other variables, this dependence is kept.<br />However, PCA can help us to see these dependencies and eliminate redundant variables.tashuhkahttps://www.blogger.com/profile/15121097471549654640noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-30703964903528446582013-04-19T23:03:08.909-04:002013-04-19T23:03:08.909-04:00Hi.how can I plot relation between LDA and PCA.Tha...Hi.how can I plot relation between LDA and PCA.Thank you.elham zamanihttps://www.blogger.com/profile/00912117655673229832noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-45761109099648348782012-12-20T12:40:39.790-05:002012-12-20T12:40:39.790-05:00how plotting a first score only .
i have a data se...how plotting a first score only .<br />i have a data set X=[x1 x2]alima chaouchhttps://www.blogger.com/profile/12070215867573873336noreply@blogger.com