Hi I'm using Weka framework to perform a data mining task. My dataset is highly imbalances. Once class consists of 1463 labels and other consists of 104. If I under sample the higher number of class becomes into 104 and the total number of variables will become 208. I feel like this is an information loss.
What will be the most suitable option to use in such cases.
You could try reweighting using the ClassBalancer filter.
This will keep all your instances, just reweight them.
Related
i hope you all are doing well!
I have a project at data mining class.Τhe data consists of numerical data and many algorithms do not work.I have to do this:"you should compare the performance of the following categorization algorithms:
RandomForest, C4.5, JRip, Bayesian Network. Where necessary use them
Weka filters to replace or create values for some properties
new properties. For comparison, adopt the Train / Test Percentage Split type with
percentage for training data equal to 80%.Describe your observations by giving tables with the results and
presenting the performance of the algorithms. Repeat the experiment by putting
percentage for training data equal to 70% and 50% presenting the results."
So my first try was to transform the data inside weka with preprocessing data numeric to nominal but a friend of mine suggest that is statistical wrong.So my second try was to use excel to transform all data even the date to numeric,remove the first row(id) and pass it to the weka(I leave double quotes only at date)
.But i have the error that i mention on the title.The dataset is:https://archive.ics.uci.edu/ml/datasets/Occupancy+Detection+
Thank you for the time.
If you define date-like data as a DATE attribute in the ARFF file (using the right format for parsing the strings), then WEKA will treat it as a numeric attribute internally (Java epoch, ie milli-seconds since 1970-01-01).
Instead of using NumericToNominal, use either the supervised or unsupervised Discretize filter if the algorithm cannot handle numeric attributes.
Converting nominal attributes to numeric ones is not a recommended approach. Instead, try the supervised or unsupervised NominalToBinary filter.
I am fairly new to machine learning and I am trying to use WEKA (GUI) to implement a neural network on a sports data set. My issue is that I want my inputs to be Arrays (each Array is a contestant with stats such as speed, winrate, etc). I am wondering how I can tell WEKA that each input is an array of values.
You can define it in an .arff file. See this website for detailed information. As the figure below.
Or after opening your data in Weka, you can convert it with the help of some filters. I do not know the current format of your data. However, if you can open it in Weka, you can edit your data with many filters. Meanwhile, artificial neural networks only accept numerical values. Among these filters, there are those who convert nominal data to numerical data. I share an image from these filters below. If you are new to this area, I recommend you to watch videos of WekaMOOC (owned by Weka developers.). I think it will be very useful. Good luck.
Weka_filters_screen
What is the minimum number of text rows needed for ground truth to do auto-labelling ? I have text file which contains 1000 rows, is this good enough to get started with auto-labelling by sagemaker ground truth ?
I'm a product manager on the Amazon SageMaker Ground Truth team, and I'm happy to help you with this question. The minimum system requirement is 1,000 objects. In practice with text classification, we typically see meaningful results (% of data auto-labeled) only once you have 2,000 to 3,000 text objects. Remember performance is variable and depends on your dataset and the complexity of your task.
From the documentation,
You should use automated data labeling only on large datasets. The neural networks used with active learning require a significant amount of data for every new dataset. With larger datasets there is more potential to automatically label the data and therefore reduce the total cost of labeling. We recommend that you use thousands of data objects when using automated data labeling. You must use at least 5,000 data objects
https://docs.aws.amazon.com/sagemaker/latest/dg/sms-automated-labeling.html
I have two datasets regarding whether a sentence contains a mention of a drug adverse event or not, both the training and test set have only two fields the text and the labels{Adverse Event, No Adverse Event} I have used weka with the stringtoWordVector filter to build a model using Random Forest on the training set.
I want to test the model built with removing the class labels from the test data set, applying the StringToWordVector filter on it and testing the model with it. When I try to do that it gives me the error saying training and test set not compatible probably because the filter identifies a different set of attributes for the test dataset. How do I fix this and output the predictions for the test set.
The easiest way to do this for a one off test is not to pre-filter the training set, but to use Weka's FilteredClassifier and configure it with the StringToWordVector filter, and your chosen classifier to do the classification. This is explained well in this video from the More Data Mining with Weka online course.
For a more general solution, if you want to build the model once then evaluate it on different test sets in future, you need to use InputMappedClassifier:
Wrapper classifier that addresses incompatible training and test data
by building a mapping between the training data that a classifier has
been built with and the incoming test instances' structure. Model
attributes that are not found in the incoming instances receive
missing values, so do incoming nominal attribute values that the
classifier has not seen before. A new classifier can be trained or an
existing one loaded from a file.
Weka requires a label even for the test data. It uses the labels or „ground truth“ of the test data to compare the result of the model against it and measure the model performance. How would you tell whether a model is performing well, if you don‘t know whether its predictions are right or wrong. Thus, the test data needs to have the very same structure as the training data in WEKA, including the labels. No worries, the labels are not used to help the model with its predictions.
The best way to go is to select cross validation (e.g. 10 fold cross validation) which automatically will split your data into 10 parts, using 9 for training and the remaining 1 for testing. This procedure is repeated 10 times so that each of the 10 parts has once been used as test data. The final performance verdict will be an average of all 10 rounds. Cross validation gives you a quite realistic estimate of the model performance on new, unseen data.
What you were trying to do, namely using the exact same data for training and testing is a bad idea, because the measured performance you end up with is way too optimistic. This means, you‘ll get very impressive figures like 98% accuracy during testing - but as soon as you use the model against new unseen data your accuracy might drop to a much worse level.
I have moderate experience with data science. I have a data set with 9500 observations and more than 4500 features most of which are highly correlated. Here is briefly what I have tried: I have dropped columns where there are less than 6000 non-NAs and have imputed NAs with their corresponding columns' median values when there are at least 6000 non-NAs. As for correlation, I have kept only features having at most 0.7 correlation with others. By doing so, I have reduced the number of features to about 750. Then I have used those features in my binary classification task in random forest.
My data set is highly unbalanced where ratio of (0:1) is (10:1). So when I apply RF with 10-fold cv, I observe too good results in each cv (AUC of 99%) which is to good to be true and in my test set I got way worse results such as 0.7. Here is my code:
import h2o
from h2o.estimators import H2ORandomForestEstimator
h2o.init(port=23, nthreads=4)
train = fs_rf[fs_rf['Year'] <= '201705']
test = fs_rf[fs_rf['Year'] > '201705']
train = train.drop('Year',axis=1)
test = test.drop('Year',axis=1)
test.head()
train = h2o.H2OFrame(train)
train['BestWorst2'] = train['BestWorst2'].asfactor()
test = h2o.H2OFrame(test)
test['BestWorst2'] = test['BestWorst2'].asfactor()
training_columns = train.drop('BestWorst2',axis=1).col_names
response_column = 'BestWorst2'
model = H2ORandomForestEstimator(ntrees=100, max_depth=20, nfolds=10, balance_classes=True)
model.train(x=training_columns, y=response_column, training_frame=train)
performance = model.model_performance(test_data=test)
print(performance)
How could I avoid this over-fitting? I have tried many different parameters in grid search but none of them improved the results.
This is not what I would call "overfitting". The reason you are seeing really good cross-validation metrics compared to your test metrics is that you have time-series data and so you can't use k-fold cross-validation to give you an accurate estimate of performance.
Performing k-fold cross-validation on a time-series dataset will give you overly-optimistic performance metrics because you are not respecting the time-series component in your data. Regular k-fold cross-validation will randomly sample from your whole dataset to create a train & validation set. Essentially, your validation strategy is "cheating" because you have "future" data included in your CV training sets (if that makes any sense).
I can see by your code that you understand that you need to train with "past" data and predict on "future" data, but if you want to read more about this topic, I'd recommend this article or this article.
One solution is to simply look at test set performance as way to evaluate your model. Another option is to use what's called "rolling" or "time-series" cross-validation, but H2O does not currently support that (though it seems like it might be added soon). Here's a ticket for this if you want to keep track of the progress.