All geek questions in one place

Search for similarities between time series in Matlab. Possible? I can not find the R-tree implementation in Matlab

I would like to implement a similarity search in matlab. I want to know if this is possible?

My plan is to use 2 popular similarity dimensions, which are Euclidean distance and dynamic time stall. Both will apply to a set of time series. My question at this point is how can I evaluate both of these measurement characteristics and accuracy? I saw some literature saying that I should use the K-NN algorithm.

Then I plan to apply dimension reduction in a set of time series. After reducing the dimension of the data set. I will need to index the dataset using the R-tree or any available indexing methods.

However, my problem is that for this I need the Mlabab R-tree code that I can hardly find on the Internet ...

I realized that most of the similarity lookup implementation is in C ++, C, and Java ... But I am not familiar with them. I hope that I can implement them in Matlab ... Any Guru could help me with this?

Also, what kind of evaluation can I do to evaluate the performance for each algorithm.

thank

+3
source share
2 answers

(R2010a, ), MATLAB k-Nearest Neighbor (kNN), KD-tree (a , R-) Toolbox. :

load fisheriris                            % Iris dataset
Q = [6 3 4 1 ; 5 4 3 2];                   % query points

% build kd-tree
knnObj = createns(meas, 'NSMethod','kdtree', 'Distance','euclidean');

% find k=5 Nearest Neighbors to Q
[idx Dist] = knnsearch(knnObj, Q, 'K',5);

.

, Image Processing Toolbox, ( ) kd- kNN. , :

[matlabroot '\images\images\private\kdtree.m']
[matlabroot '\images\images\private\nnsearch.m']

( ), ; /, , , , kNN, . , , / ..

+3

Source: https://habr.com/ru/post/1756347/

More articles:


All Articles