Re: K Means Clustering Explanation
Thanks Alessandro and Christoph. I appreciate the feedback, but I'm still having issues determining how to actually accomplish this with the API. Can anyone point me to an example in code showing how to accomplish this? On Fri, Mar 2, 2018 2:37 AM, Alessandro Solimando alessandro.solima...@gmail.com wrote: Hi Matt,similarly to what Christoph does, I first derive the cluster id for the elements of my original dataset, and then I use a classification algorithm (cluster ids being the classes here). For this method to be useful you need a "human-readable" model, tree-based models are generally a good choice (e.g., Decision Tree). However, since those models tend to be verbose, you still need a way to summarize them to facilitate readability (there must be some literature on this topic, although I have no pointers to provide). Hth,Alessandro On 1 March 2018 at 21:59, Christoph Brücke <carabo...@gmail.com> wrote: Hi Matt, I see. You could use the trained model to predict the cluster id for each training point. Now you should be able to create a dataset with your original input data and the associated cluster id for each data point in the input data. Now you can group this dataset by cluster id and aggregate over the original 5 features. E.g., get the mean for numerical data or the value that occurs the most for categorical data. The exact aggregation is use-case dependent. I hope this helps,Christoph Am 01.03.2018 21:40 schrieb "Matt Hicks" <m...@outr.com>: Thanks for the response Christoph. I'm converting large amounts of data into clustering training and I'm just having a hard time reasoning about reversing the clusters (in code) back to the original format to properly understand the dominant values in each cluster. For example, if I have five fields of data and I've trained ten clusters of data I'd like to output the five fields of data as represented by each of the ten clusters. On Thu, Mar 1, 2018 2:36 PM, Christoph Brücke carabo...@gmail.com wrote: Hi matt, the cluster are defined by there centroids / cluster centers. All the points belonging to a certain cluster are closer to its than to the centroids of any other cluster. What I typically do is to convert the cluster centers back to the original input format or of that is not possible use the point nearest to the cluster center and use this as a representation of the whole cluster. Can you be a little bit more specific about your use-case? Best,Christoph Am 01.03.2018 20:53 schrieb "Matt Hicks" <m...@outr.com>: I'm using K Means clustering for a project right now, and it's working very well. However, I'd like to determine from the clusters what information distinctions define each cluster so I can explain the "reasons" data fits into a specific cluster. Is there a proper way to do this in Spark ML?
K Means Clustering Explanation
I'm using K Means clustering for a project right now, and it's working very well. However, I'd like to determine from the clusters what information distinctions define each cluster so I can explain the "reasons" data fits into a specific cluster. Is there a proper way to do this in Spark ML?
Re: [Spark ML] Positive-Only Training Classification in Scala
If I try to use LogisticRegression with only positive training it always gives me positive results: Positive Only private def positiveOnly(): Unit = {val training = spark.createDataFrame(Seq( (1.0, Vectors.dense(0.0, 1.1, 0.1)), (1.0, Vectors.dense(0.0, 1.0, -1.0)), (1.0, Vectors.dense(0.2, 1.3, 1.0)), (1.0, Vectors.dense(0.1, 1.2, -0.5)))).toDF("label", "features")val lr = new LogisticRegression() lr.setMaxIter(10).setRegParam(0.01)val model = lr.fit(training)val test = spark.createDataFrame(Seq( (1.0, Vectors.dense(-1.0, 1.5, 1.3)), (0.0, Vectors.dense(3.0, 2.0, -0.1)), (1.0, Vectors.dense(0.0, 2.2, -1.5)) )).toDF("label", "features")model.transform(test) .select("features", "label", "probability", "prediction") .collect() .foreach { case Row(features: Vector, label: Double, prob: Vector, prediction: Double) =>println(s"($features, $label) -> prob=$prob, prediction=$prediction") } } Not using Mixmax yet? The results look like this: [info] ([-1.0,1.5,1.3], 1.0) -> prob=[0.0,1.0], prediction=1.0[info] ([3.0,2.0,-0.1], 0.0) -> prob=[0.0,1.0], prediction=1.0[info] ([0.0,2.2,-1.5], 1.0) -> prob=[0.0,1.0], prediction=1.0 On Tue, Jan 16, 2018 8:51 AM, Matt Hicks m...@outr.com wrote: Hi Hari, I'm not sure I understand. I apologize, I'm still pretty new to Spark and Spark ML. Can you point me to some example code or documentation that would more fully represent this? Thanks On Tue, Jan 16, 2018 2:54 AM, hosur narahari hnr1...@gmail.com wrote: You can make use of probability vector from spark classification.When you run spark classification model for prediction, along with classifying into its class spark also gives probability vector(what's the probability that this could belong to each individual class) . So just take the probability corresponding to the donor class. And it'll be same as what's the probability the a person will become donor. Best Regards,Hari On 15 Jan 2018 11:51 p.m., "Matt Hicks" <m...@outr.com> wrote: I'm attempting to create a training classification, but only have positive information. Specifically in this case it is a donor list of users, but I want to use it as training in order to determine classification for new contacts to give probabilities that they will donate. Any insights or links are appreciated. I've gone through the documentation but have been unable to find any references to how I might do this. Thanks --- Matt Hicks Chief Technology Officer 405.283.6887 | http://outr.com
Re: [Spark ML] Positive-Only Training Classification in Scala
Hi Hari, I'm not sure I understand. I apologize, I'm still pretty new to Spark and Spark ML. Can you point me to some example code or documentation that would more fully represent this? Thanks On Tue, Jan 16, 2018 2:54 AM, hosur narahari hnr1...@gmail.com wrote: You can make use of probability vector from spark classification.When you run spark classification model for prediction, along with classifying into its class spark also gives probability vector(what's the probability that this could belong to each individual class) . So just take the probability corresponding to the donor class. And it'll be same as what's the probability the a person will become donor. Best Regards,Hari On 15 Jan 2018 11:51 p.m., "Matt Hicks" <m...@outr.com> wrote: I'm attempting to create a training classification, but only have positive information. Specifically in this case it is a donor list of users, but I want to use it as training in order to determine classification for new contacts to give probabilities that they will donate. Any insights or links are appreciated. I've gone through the documentation but have been unable to find any references to how I might do this. Thanks --- Matt Hicks Chief Technology Officer 405.283.6887 | http://outr.com
Re: [Spark ML] Positive-Only Training Classification in Scala
Is it fair to assume this is what I need? https://github.com/ispras/pu4spark On Mon, Jan 15, 2018 1:55 PM, Georg Heiler georg.kf.hei...@gmail.com wrote: As far as I know spark does not implement such algorithms. In case the dataset is small http://scikit-learn.org/stable/modules/generated/sklearn.svm.OneClassSVM.html might be of interest to you. Jörn Franke <jornfra...@gmail.com> schrieb am Mo., 15. Jan. 2018 um 20:04 Uhr: I think you look more for algorithms for unsupervised learning, eg clustering. Depending on the characteristics different clusters might be created , eg donor or non-donor. Most likely you may find also more clusters (eg would donate but has a disease preventing it or too old). You can verify which clusters make sense for your approach so I recommend not only try two clusters but multiple and see which number is more statistically significant . On 15. Jan 2018, at 19:21, Matt Hicks <m...@outr.com> wrote: I'm attempting to create a training classification, but only have positive information. Specifically in this case it is a donor list of users, but I want to use it as training in order to determine classification for new contacts to give probabilities that they will donate. Any insights or links are appreciated. I've gone through the documentation but have been unable to find any references to how I might do this. Thanks --- Matt Hicks Chief Technology Officer 405.283.6887 | http://outr.com
[Spark ML] Positive-Only Training Classification in Scala
I'm attempting to create a training classification, but only have positive information. Specifically in this case it is a donor list of users, but I want to use it as training in order to determine classification for new contacts to give probabilities that they will donate. Any insights or links are appreciated. I've gone through the documentation but have been unable to find any references to how I might do this. Thanks --- Matt Hicks Chief Technology Officer 405.283.6887 | http://outr.com