data-mining

I have a dataset of about 100,000 records about buying pattern of customers. The data set contains Age (continuous value from 2 to 120) but I have plan also to categorize into age ranges. Gender (either 0 or 1) Address (can be only six types or I can also represent using numbers from 1 to 6) Preference shop (can be from only 7 shops) which is my class problem. So my problem is to classify and predict the customers based on their Age,gender and location for Preference shop. I have tried to use naive and decision trees but their classification accuracy is little bit low below. I am thinking also

This question already has answers here : What is an intuitive explanation of the Expectation Maximization technique? [closed] (8 answers) Could anyone provide a simple numeric example of the EM algorithm as I am not sure about the formulas given? A really simple one with 4 or 5 Cartesian coordinates would perfectly do. what about this: http://en.wikibooks.org/wiki/Data_Mining_Algorithms_In_R/Clustering/Expectation_Maximization_(EM)#A_simple_example I had also written a simple example in (edit)R a year ago, unfortunately I am unable to locate it. I'll try again to find it later. EDIT: Here it

I'm writting a piece of code to evaluate my Clustering Algorithm and I find that every kind of evaluation method needs the basic data from a m*n matrix like A = {aij} where aij is the number of data points that are members of class ci and elements of cluster kj . But there appear to be two of this type of matrix in Introduction to Data Mining (Pang-Ning Tan et al.), one is the Confusion Matrix, the other is the Contingency Table. I do not fully understand the difference between the two. Which best describes the matrix I want to use? SpeedBirdNine Wikipedia's definition : In the field of

I am a freshman in LDA and I want to use it in my work. However, some problems appear. In order to get the best performance, I want to estimate the best topic number. After reading "Finding Scientific topics", I know that I can calculate logP(w|z) firstly and then use the harmonic mean of a series of P(w|z) to estimate P(w|T). My question is what does the "a series of" mean? Unfortunately, there is no hard science yielding the correct answer to your question. To the best of my knowledge, hierarchical dirichlet process (HDP) is quite possibly the best way to arrive at the optimal number of

I'm working with some large transactions data. I've been using read.transactions and apriori (parts of the arules package) to mine for frequent item pairings. My problem is this: when rules are generated (using "inspect()") I can easily view them in the R console. Right now I'm manually copying the results into a text file, then saving and opening in excel. I'd like to just save the generated rules using write.csv, or something similar, but when I try, I receive an error that the data cannot be coerced into data.frame. Does anyone have experience doing this successfully in R? I know I'm

I am evaluating datamining packages. I have find these two so far: RapidMiner Weka Do you have any experience to share with these two products, or any other product to recommend me? Thanks According to the yearly KDnuggets Polls 2007, 2008, and 2009, RapidMiner is the most widely used Open Source Data Mining Solution among data mining experts world-wide: KDnuggets Data Mining Tool Poll 2009 RapidMiner is open source and 100% Java, RapidMiner is much more flexible and offers significantly more functionality than Weka and KNIME. Regarding SVM implementations: Weka comes with one such