Imbalanced classification: order of oversampling vs. scaling features? - machine-learning

When performing classification (for example, logistic regression) with an imbalanced dataset (e.g., fraud detection), is it best to scale/zscore/standardize the features before over-sampling the minority class, or to balance the classes before scaling features?
Secondly, does the order of these steps affect how features will eventually be interpreted (when using all data, scaled+balanced, to train a final model)?
Here's an example:
Scale first:
Split data into train/test folds
Calculate mean/std using all training (imbalanced) data; scale the training data using these calculations
Oversample minority class in the training data (e.g, using SMOTE)
Fit logistic regression model to training data
Use mean/std calculations to scale the test data
Predict class with imbalanced test data; assess acc/recall/precision/auc
Oversample first
Split data into train/test folds
Oversample minority class in the training data (e.g, using SMOTE)
Calculate mean/std using balanced training data; scale the training data using these calculations
Fit logistic regression model to training data
Use mean/std calculations to scale the test data
Predict class with imbalanced test data; assess acc/recall/precision/auc

Since under- and over-sampling techniques often depend on k-NN or k-means-like algorithms which use the Euclidean distance between data points, it is safer to scale before resampling. However, in real-life, the order hardly matters.

You may have meant it implicitly, but you need to apply the mean/std to scale the training data as well, and that needs to happen before you fit the model.
Barring that point, there isn't a definitive answer on this. The best thing would be to simply try both and see which works best for your data.
For you own understanding of the model on the resulting data, you may want to instead play with computing the mean and standard deviation of the minority and majority classes. If they have similar statistics, then we wouldn't expect much of a difference between scale first or over-sample first.
If the means and standard deviations are very different, the results may differ significantly. But that may also mean the problem has greater separation, and you may expect a higher classification accuracy.

Related

How can we modify the training dataset to achieve better accuracy on the minority class for any classification algorithm?

In many training datasets, the class distribution is skewed, I.e. one class is very frequent (e.g. 95%), while another is rare (e.g. 5%). In such applications, the Naive classifier that assigns all test cases to the majority class achieves very high accuracy. However, the Naive classifier does not predict any cases of the minority class, although it is often more important than the majority class.
So are there any ways to improve the accuracy on the minority class just by modifying the training dataset? or Do we have to modify the classification algorithm as well?
This is the issue of the Imbalanced classification. Below are the methods used to treat the Imbalanced Datasets.
1.Undersampling
2.Oversampling
3.Synthetic Data Generation
4.Cost-Sensitive Learning
1.Undersampling
This method works with the majority class. It reduces the number of observations from the majority class to make the data set balanced. This method is best to use when the data set is huge and reducing the number of training samples helps to improve run time and storage troubles.
Undersampling methods are of 2 types: Random and Informative.
Oversampling
This method works with minority class. It replicates the observations from minority class to balance the data. It is also known as upsampling. Similar to undersampling, this method also can be divided into two types: Random Oversampling and Informative Oversampling.
Synthetic Data Generation
In simple words, instead of replicating and adding observations from the minority class, it overcomes imbalances by generates artificial data. It is also a type of oversampling technique.
In regards to synthetic data generation, synthetic minority oversampling technique (SMOTE) is a powerful and widely used method. SMOTE algorithm creates artificial data based on feature space (rather than data space) similarities from minority samples. We can also say, it generates a random set of minority class observations to shift the classifier learning bias towards minority class.
Cost-Sensitive Learning (CSL)
It is another commonly used method to handle classification problems with imbalanced data. It’s an interesting method. In simple words, this method evaluates the cost associated with misclassifying observations.
It does not create a balanced data distribution. Instead, it highlights the imbalanced learning problem by using cost matrices which describe the cost for misclassification in a particular scenario. Recent researches have shown that cost-sensitive learning has many times outperformed sampling methods. Therefore, this method provides a likely alternative to sampling methods.
To build these in R refer the link https://www.analyticsvidhya.com/blog/2016/03/practical-guide-deal-imbalanced-classification-problems/
There are many ways. You can upsample miority class, downsample majority, do some weighting based on loss, or some other measures. As you've tagged your post to be related to TF I suggest you look at Importance Sampling for Keras, http://idiap.ch/~katharas/importance-sampling/ it has some ready to use wrappers that you put around your model.

Data Science Scaling/Normalization real case

When do data pre-processing, it is suggested to do either scaling or normalization. It is easy to do it when you have data on your hand. You have all the data and can do it right away. But after the model built and run, does the first data that comes in need to be scaled or normalized? If it needed, it only one single row how to scale or normalize it? How do we know what is the min/max/mean/stdev from each feature? And how is the incoming data is the min/max/mean each feature?
Please advise
First of all you should know when to use scaling and normalization.
Scaling - scaling is nothing but to transform your features to comparable magnitudes.Let say if you have features like person's income and you noticed that some have value of order 10^3 and some have 10^6.Now if you model your problem with this features then algorithms like KNN, Ridge Regression will give higher weight to higher magnitude of such attributes.To prevent this you need to first scale your features.Min-Max scaler is one of the most used scaling.
Mean Normalisation -
If after examining the distribution of the feature and you found that feature is not centered around zero then for the algorithm like svm where objective function already assumes zero mean and same order variance, we could have problem in modeling.So here you should do Mean Normalisation.
Standardization - For the algorithm like svm, neural network, logistic regression it is necessary to have a variance of the feature in the same order.So why don't we make it to one.So in standardization, we make the distribution of features to zero mean and unit variance.
Now let's try to answer your question in terms of training and testing set.
So let's say you are training your model on 50k dataset and testing on 10k dataset.
For the above three transformations, the standard approach says that you should fit any normalizer or scaler to only training dataset and use only transform for the testing dataset.
In our case, if we want to use standardization then we will first fit our standardizer on 50k training dataset and then used to transform it 50k training dataset and also testing dataset.
Note - We shouldn't fit our standardizer to test dataset, in place of we will use already fitted standardizer to transform testing dataset.
Yes, you need to apply normalization to the input data, else the model will predict nonsense.
You also have to save the normalization coefficients that were used during training, or from training data. Then you have to apply the same coefficients to incoming data.
For example if you use min-max normalization:
f_n = (f - min(f)) / (max(f) - min_(f))
Then you need to save the min(f) and max(f) in order to perform normalization for new data.

Machine Learning Experiment Design with Small Positive Sample Set in Sci-kit Learn

I am interested in any tips on how to train a set with a very limited positive set and a large negative set.
I have about 40 positive examples (quite lengthy articles about a particular topic), and about 19,000 negative samples (most drawn from the sci-kit learn newsgroups dataset). I also have about 1,000,000 tweets that I could work with.. negative about the topic I am trying to train on. Is the size of the negative set versus the positive going to negatively influence training a classifier?
I would like to use cross-validation in sci-kit learn. Do I need to break this into train / test-dev / test sets? Is know there are some pre-built libraries in sci-kit. Any implementation examples that you recommend or have used previously would be helpful.
Thanks!
The answer to your first question is yes, the amount by which it will affect your results depends on the algorithm. My advive would be to keep an eye on the class-based statistics such as recall and precision (found in classification_report).
For RandomForest() you can look at this thread which discusses
the sample weight parameter. In general sample_weight is what
you're looking for in scikit-learn.
For SVM's have a look at either this example or this
example.
For NB classifiers, this should be handled implicitly by Bayes
rule, however in practice you may see some poor performances.
For you second question it's up for discussion, personally I break my data into a training and test split, perform cross validation on the training set for parameter estimation, retrain on all the training data and then test on my test set. However the amount of data you have may influence the way you split your data (more data means more options).
You could probably use Random Forest for your classification problem. There are basically 3 parameters to deal with data imbalance. Class Weight, Samplesize and Cutoff.
Class Weight-The higher the weight a class is given, the more its error rate is decreased.
Samplesize- Oversample the minority class to improve class imbalance while sampling the defects for each tree[not sure if Sci-kit supports this, used to be param in R)
Cutoff- If >x% trees vote for the minority class, classify it as minority class. By default x is 1/2 in Random forest for 2-class problem. You can set it to a lower value for the minority class.
Check out balancing predict error at https://www.stat.berkeley.edu/~breiman/RandomForests/cc_home.htm
For the 2nd question if you are using Random Forest, you do not need to keep separate train/validation/test set. Random Forest does not choose any parameters based on a validation set, so validation set is un-necessary.
Also during the training of Random Forest, the data for training each individual tree is obtained by sampling by replacement from the training data, thus each training sample is not used for roughly 1/3 of the trees. We can use the votes of these 1/3 trees to predict the out of box probability of the Random forest classification. Thus with OOB accuracy you just need a training set, and not validation or test data to predict performance on unseen data. Check Out of Bag error at https://www.stat.berkeley.edu/~breiman/RandomForests/cc_home.htm for further study.

Training on imbalanced data using TensorFlow

The Situation:
I am wondering how to use TensorFlow optimally when my training data is imbalanced in label distribution between 2 labels. For instance, suppose the MNIST tutorial is simplified to only distinguish between 1's and 0's, where all images available to us are either 1's or 0's. This is straightforward to train using the provided TensorFlow tutorials when we have roughly 50% of each type of image to train and test on. But what about the case where 90% of the images available in our data are 0's and only 10% are 1's? I observe that in this case, TensorFlow routinely predicts my entire test set to be 0's, achieving an accuracy of a meaningless 90%.
One strategy I have used to some success is to pick random batches for training that do have an even distribution of 0's and 1's. This approach ensures that I can still use all of my training data and produced decent results, with less than 90% accuracy, but a much more useful classifier. Since accuracy is somewhat useless to me in this case, my metric of choice is typically area under the ROC curve (AUROC), and this produces a result respectably higher than .50.
Questions:
(1) Is the strategy I have described an accepted or optimal way of training on imbalanced data, or is there one that might work better?
(2) Since the accuracy metric is not as useful in the case of imbalanced data, is there another metric that can be maximized by altering the cost function? I can certainly calculate AUROC post-training, but can I train in such a way as to maximize AUROC?
(3) Is there some other alteration I can make to my cost function to improve my results for imbalanced data? Currently, I am using a default suggestion given in TensorFlow tutorials:
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(pred, y))
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)
I have heard this may be possible by up-weighting the cost of miscategorizing the smaller label class, but I am unsure of how to do this.
(1)It's ok to use your strategy. I'm working with imbalanced data as well, which I try to use down-sampling and up-sampling methods first to make the training set even distributed. Or using ensemble method to train each classifier with an even distributed subset.
(2)I haven't seen any method to maximise the AUROC. My thought is that AUROC is based on true positive and false positive rate, which doesn't tell how well it works on each instance. Thus, it may not necessarily maximise the capability to separate the classes.
(3)Regarding weighting the cost by the ratio of class instances, it similar to Loss function for class imbalanced binary classifier in Tensor flow
and the answer.
Regarding imbalanced datasets, the first two methods that come to mind are (upweighting positive samples, sampling to achieve balanced batch distributions).
Upweighting positive samples
This refers to increasing the losses of misclassified positive samples when training on datasets that have much fewer positive samples. This incentivizes the ML algorithm to learn parameters that are better for positive samples. For binary classification, there is a simple API in tensorflow that achieves this. See (weighted_cross_entropy) referenced below
https://www.tensorflow.org/api_docs/python/tf/nn/weighted_cross_entropy_with_logits
Batch Sampling
This involves sampling the dataset so that each batch of training data has an even distribution positive samples to negative samples. This can be done using the rejections sampling API provided from tensorflow.
https://www.tensorflow.org/api_docs/python/tf/contrib/training/rejection_sample
I'm one who struggling with imbalanced data. What my strategy to counter imbalanced data are as below.
1) Use cost function calculating 0 and 1 labels at the same time like below.
cost = tf.reduce_mean(-tf.reduce_sum(y*tf.log(_pred) + (1-y)*tf.log(1-_pred), reduction_indices=1))
2) Use SMOTE, oversampling method making number of 0 and 1 labels similar. Refer to here, http://comments.gmane.org/gmane.comp.python.scikit-learn/5278
Both strategy worked when I tried to make credit rating model.
Logistic regression is typical method to handle imbalanced data and binary classification such as predicting default rate. AUROC is one of the best metric to counter imbalanced data.
1) Yes. This is well received strategy to counter imbalanced data. But this strategy is good in Neural Nets only if you using SGD.
Another easy way to balance the training data is using weighted examples. Just amplify the per-instance loss by a larger weight/smaller when seeing imbalanced examples. If you use online gradient descent, it can be as simple as using a larger/smaller learning rate when seeing imbalanced examples.
Not sure about 2.

What's the difference between ANN, SVM and KNN classifiers?

I am doing remote sensing image classification. I am using the object-oriented method: first I segmented the image to different regions, then I extract the features from regions such as color, shape and texture. The number of all features in a region may be 30 and commonly there are 2000 regions in all, and I will choose 5 classes with 15 samples for every class.
In summary:
Sample data 1530
Test data 197530
How do I choose the proper classifier? If there are 3 classifiers (ANN, SVM, and KNN), which should I choose for better classification?
KNN is the most basic machine learning algorithm to paramtise and implement, but as alluded to by #etov, would likely be outperformed by SVM due to the small training data sizes. ANNs have been observed to be limited by insufficient training data also. However, KNN makes the least number of assumptions regarding your data, other than that accurate training data should form relatively discrete clusters. ANN and SVM are notoriously difficult to paramtise, especially if you wish to repeat the process using multiple datasets and rely upon certain assumptions, such as that your data is linearly separable (SVM).
I would also recommend the Random Forests algorithm as this is easy to implement and is relatively insensitive to training data size, but I would advise against using very small training data sizes.
The scikit-learn module contains these algorithms and is able to cope with large training data sizes, so you could increase the number of training data samples. the best way to know for sure would be to investigate them yourself, as suggested by #etov
If your "sample data" is the train set, it seems very small. I'd first suggest using more than 15 examples per class.
As said in the comments, it's best to match the algorithm to the problem, so you can simply test to see which algorithm works better. But to start with, I'd suggest SVM: it works better than KNN with small train sets, and generally easier to train then ANN, as there are less choices to make.
Have a look at below mind map
KNN: KNN performs well when sample size < 100K records, for non textual data. If accuracy is not high, immediately move to SVC ( Support Vector Classifier of SVM)
SVM: When sample size > 100K records, go for SVM with SGDClassifier.
ANN: ANN has evolved overtime and they are powerful. You can use both ANN and SVM in combination to classify images
More details are available #semanticscholar.org

Resources