I am trying to understand PCA and K-Means algorithms in order to extract some relevant features from a set of features.
I don't know what branch of computer science study these topics, seems on internet there aren't good resources, just some paper that I don't understand well. An example of paper http://www.ifp.illinois.edu/~qitian/e_paper/icip02/icip02.pdf
I have csv files of pepole walks composed as follow:
TIME, X, Y, Z, these values are registred by the accelerometer
What I did
I transformed the dataset as a table in Python
I used tsfresh, a Python library, to extract from each walk a vector of features, these features are a lot, 2k+ features from each walk.
I have to use PFA, Principal Feature Analysis, to select the relevant features from the set of
vectors features
In order to do the last point, I have to reduce the dimension of the set of features walks with PCA (PCA will make the data different from the original one cause it modifies the data with the eigenvectors and eigenvalues of the covariance matrix of the original data). Here I have the first question:
How the input of PCA should look? The rows are the number of walks and the columns are the features or viceversa, so the rows are the number of the features and the columns are the number of walks of pepole?
After I reduced this data, I should use the K-Means algorithm on the reduced 'features' data. How the input should look in the K-Means? And what's the propouse on using this algorithm? All I know this algorithm it's used to 'cluster' some data, so in each cluster there are some 'points' based on some rule. What I did and think is:
If I use in PCA an input that looks like: the rows are the number of walks and the columns are the number of features, then for K-Means I should change the columns with rows cause in this way each point it's a feature (but this is not the original data with the features, it's just the reduced one, so I don't know). So then for each cluster I see with euclidean distance who has the lower distance from the centroid and select that feature. So how many clusters I should declare? If I declare that the clusters are the same as the number of features, I will extract always the same number of features. How can I say that a point in the reduced data correspond to this feature in the original set of features?
I know it's not correct what I am saying maybe, but I am trying to understand it, can some of you help me? If am I in the right way? Thanks!
For the PCA, make sure you separate the understanding of the method the algorithm uses (eigenvectors and such) and the result. The result, is a linear mapping, mapping the original space A, to A', where possibly, the dimension (number of features in your case) is less than the original space A.
So the first feature/element in space A', is a linear combination of features of A.
The row/column depends on implementation, but if you use scikit PCA the columns are the features.
You can feed the PCA output, the A' space, to K-means, and it will cluster them, based on a space of usually reduced dimension.
Each point will be part of a cluster, and the idea is that if you would calculate K-Means on A, you would probably end up with the same/similar clusters like with A'. Computationally A' is a lot cheaper. You now have a clustering, on A' and A. As we agree that points similar in A' are also similar in A.
The number of clusters is difficult to answer, if you don't know anything search the elbow method. But say you want to get a sense of different type of things you have, I argue go for 3~8 and not too much, compare 2-3 points closest to
each center, and you have something consumable. The number of features can be larger than the number of clusters. e.g. If we want to know the most dense area in some area (2D) you can easily have 50 clusters, to get a sense where 50 cities could be. Here we have number of cluster way higher than space dimension, and it makes sense.
Related
I'm not a trained statistician so I apologize for the incorrect usage of some words. I'm just trying to get some good results from the Weka Nearest Neighbor algorithms. I'll use some redundancy in my explanation as a means to try to get the concept across:
Is there a way to normalize a multi-dimensional space so that the distances between any two instances are always proportional to the effect on the dependent variable?
In other words I have a statistical data set and I want to use a "nearest neighbor" algorithm to find instances that are most similar to a specified test instance. Unfortunately my initial results are useless because two attributes that are very close in value weakly correlated to the dependent variable would incorrectly bias the distance calculation.
For example let's say you're trying to find the nearest-neighbor of a given car based on a database of cars: make, model, year, color, engine size, number of doors. We know intuitively that the make, model, and year have a bigger effect on price than the number of doors. So a car with identical color, door count, may not be the nearest neighbor to a car with different color/doors but same make/model/year. What algorithm(s) can be used to appropriately set the weights of each independent variable in the Nearest Neighbor distance calculation so that the distance will be statistically proportional (correlated, whatever) to the dependent variable?
Application: This can be used for a more accurate "show me products similar to this other product" on shopping websites. Back to the car example, this would have cars of same make and model bubbling up to the top, with year used as a tie-breaker, and then within cars of the same year, it might sort the ones with the same number of cylinders (4 or 6) ahead of the ones with the same number of doors (2 or 4). I'm looking for an algorithmic way to derive something similar to the weights that I know intuitively (make >> model >> year >> engine >> doors) and actually assign numerical values to them to be used in the nearest-neighbor search for similar cars.
A more specific example:
Data set:
Blue,Honda,6-cylinder
Green,Toyota,4-cylinder
Blue,BMW,4-cylinder
now find cars similar to:
Blue,Honda,4-cylinder
in this limited example, it would match the Green,Toyota,4-cylinder ahead of the Blue,Honda,6-cylinder because the two brands are statistically almost interchangeable and cylinder is a stronger determinant of price rather than color. BMW would match lower because that brand tends to double the price, i.e. placing the item a larger distance.
Final note: the prices are available during training of the algorithm, but not during calculation.
Possible you should look at Solr/Lucene for this aim. Solr provides a similarity search based field value frequency and it already has functionality MoreLikeThis for find similar items.
Maybe nearest neighbor is not a good algorithm for this case? As you want to classify discrete values it can become quite hard to define reasonable distances. I think an C4.5-like algorithm may better suit the application you describe. On each step the algorithm would optimize the information entropy, thus you will always select the feature that gives you the most information.
Found something in the IEEE website. The algorithm is called DKNDAW ("dynamic k-nearest-neighbor with distance and attribute weighted"). I couldn't locate the actual paper (probably needs a paid subscription). This looks very promising assuming that the attribute weights are computed by the algorithm itself.
I am looking for clustering algorithm which can handle with multiple time series information for each objects.
For example, for company "A" we have time series of 3 features(ex. income, sales, inventory)
At the same way, company "B" also has same time series of same features. and so on..
Then, how we can make cluster between set of company?
Is there some wise way to handle this?
A lot of clustering algorithms ask you to provide some measure of the similarity or distance between two points. It is really up to you to decide what features are important and what the distance really is. One way forwards would be to use the correlation between two time series. This gives you a similarity. If you have to convert this to a distance I would use sqrt(1-r), where r is the correlation, because if you look e.g. at the equation at the bottom of http://www.analytictech.com/mb876/handouts/distance_and_correlation.htm you can see that this is proportional to a distance if you have points in n-dimensional space. If you have three different time series (income, sales, inventory) I would use the sum of the three distances worked out from the correlations between the two time series of the same type.
Another option, especially if the time series are not very long, would be to regard a time series of length n as a point in n-dimensional space and feed this into the clustering algorithm, or use http://en.wikipedia.org/wiki/Principal_component_analysis to reduce the n dimensions down to 1 by looking at the most significant components (while you are doing this, it never hurts to plot the points using the least significant components and investigate points that stand out from the others. Points where the data is in error sometimes stand out here).
I have a huge journal with actions done by users (like, for example, moderating contents).
I would like to find the 'mass' actions, meaning the actions that are too dense (the user probably made those actions without thinking it too much :) ).
That would translate to clustering the actions by date (in a linear space), and to marking the clusters that are too dense.
I am no expert in clustering algorithms and methods, but I think the k-means clustering would not do the trick, since I don't know the number of clusters.
Also, ideally, I would also like to 'fine tune' the algorithm.
What would you advice?
P.S. Here are some resources that I found (in Ruby):
hierclust - a simple hierarchical clustering library for spatial data
AI4R - library that implements some clustering algorithms
K-means would probably do a good job as long as you're interested in an a priori known number of clusters. Since you don't you might consider reading about the LBG algorithm, which is based on k-means and is used in data compression for vector quantisation. It's basically iterative k-means which splits centroids after they converge and keeps splitting until you achieve an acceptable number of clusters.
On the other hand, since your data is one-dimensional, you could do something completely different.
Assume that you've got actions which took place at 5 points in time: (8, 11, 15, 16, 17). Let's plot a Gaussian for each of these actions with μ equal to the time and σ = 3.
Now let's see how a sum of values of these Gaussians looks like.
It shows a density of actions with a peak around 16.
Based on this observation I propose a following simple algorithm.
Create a vector of zeroes for the time range of interest.
For each action calculate the Gaussian and add it to the vector.
Scan the vector looking for values which are greater than the maximum value in the vector multiplied by α.
Note that for each action only a small section of the vector needs updates because values of a Gaussian converge to zero very quickly.
You can tune the algorithm by adjusting values of
α ∈ [0,1], which indicates how significant a peak of activity has to be to be noted,
σ, which affects the distance of actions which are considered close to each other, and
time periods per vector's element (minutes, seconds, etc.).
Notice that the algorithm is linear with regard to the number of actions. Moreover, it shouldn't be difficult to parallelise: split your data across multiple processes summing Gaussians and then sum generated vectors.
Have a look at density based clustering. E.g. DBSCAN and OPTICS.
This sounds like exactly what you want.
Looking for an algorithm that can be used to determine groups of units that move together as a squad in a real time strategy game like StarCraft. The direction that I am currently look at is a clustering algorithm but having a hard time finding which one would work best since units are moving as a group not just standing still. Any help would be great.
K-means is not the best choice, as it requires you to specify the number of clusters you expect to find. Some might contain single objects then.
I recommend adapting DBSCAN. In particular, the generalized version GDBSCAN.
For this, you need to define what constitutes the neighborhood of a unit - say, any other unit within a range of 2 that is belonging to the same player and moving approximately in the same direction (up to a certain delta threshold in x and y velocity).
Next, you need to specify when you consider units to start forming an initial cluster, called "core point". Say that is a minimum of 3 units.
Then using DBSCAN is quite basic, and should give you good results. You need to fine-tune the parameters a bit. Things like this minimum size are clearly an input parameter, and depend on your use case. So is the neighborhood definition: you are looking for groups that move into the same direction, this information needs to be put into the algorithm somehow. With GDBSCAN this is trivial, by adjusting the neighborhood definition.
You may want to look at a number of classification algorithms, like k-Nearest Neighbor or Support Vector Machines
Kmeans algorithm is quite simple and standard approach. You can check if it works:
Assume a group of data points, such as one plotted here (this graph isn't specific to my problem, but just used as a suitable example):
Inspecting the scatter graph visually, it's fairly obvious the data points form two 'groups', with some random points that do not obviously belong to either.
I'm looking for an algorithm, that would allow me to:
start with a data set of two or more dimensions.
detect such groups from the dataset without prior knowledge on how many (or if any) might be there
once the groups have been detected, 'ask' the model of groups, if a new sample point seems to fit to any of the groups
There are many choices, but if you are interested in the probability that a new data point belongs to a particular mixture, I would use a probabilistic approach such as Gaussian mixture modeling either estimated by maximum likelihood or Bayes.
Maximum likelihood estimation of mixtures models is implemented in Matlab.
Your requirement that the number of components is unknown makes your model more complex. The dominant probabilistic approach is to place a Dirichlet Process prior on the mixture distribution and estimate by some Bayesian method. For instance, see this paper on infinite Gaussian mixture models. The DP mixture model will give you inference over the number of components and the components each elements belong to, which is exactly what you want. Alternatively you could perform model selection on the number of components, but this is generally less elegant.
There are many implementation of DP mixture models models, but they may not be as convenient. For instance, here's a Matlab implementation.
Your graph suggests you are an R user. In that case, if you are looking for prepacked solutions, the answer to your question lies on this Task View for cluster analysis.
I think you are looking for something along the lines of a k-means clustering algorithm.
You should be able to find adequate implementations in most general purpose languages.
You need one of clustering algorithms. All of them can be devided in 2 groups:
you specify number of groups (clusters) - 2 clusters in your example
algorithm try to guess correct number of clusters by itself
If you want algorithm of 1st type then K-Means is what you really need.
If you want algorithm of 2nd type then you probably need one of hierarchical clustering algorithms. I haven't ever implement any of them. But I see an easy way to improve K-means in such way thay it will be unnecessary to specify number of clusters.