Visual Guides to CRISP-DM ,KDD and SEMMA

UPDATED- Here are three great examples of a visualization making a process easy to understand. Please click on the images to read them clearly.

1) It visualizes CRISP-DM and is made by Nicole Leaper (http://exde.wordpress.com/2009/03/13/a-visual-guide-to-crisp-dm-methodology/)

12345

2) KDD -Knowledge Discovery in Databases -visualization by Fayyad whom I have interviewed here at http://www.decisionstats.com/interview-dr-usama-fayyad-founder-open-insights-llc/

and work By Gregory Piatetsky Shapiro interviewed by this website here

http://decisionstats.com/2009/08/13/interview-gregory-piatetsky-kdnuggets-com/

kdd

3) I am also attaching a visual representation of SEMMA from http://www.dataprix.net/en/blogs/respinosamilla/theory-data-mining

metodo-semma

 

Data Frame in Python

Exploring some Python Packages and R packages to move /work with both Python and R without melting your brain or exceeding your project deadline

—————————————

If you liked the data.frame structure in R, you have some way to work with them at a faster processing speed in Python.

Here are three packages that enable you to do so-

(1) pydataframe http://code.google.com/p/pydataframe/

An implemention of an almost R like DataFrame object. (install via Pypi/Pip: “pip install pydataframe”)

Usage:

        u = DataFrame( { "Field1": [1, 2, 3],
                        "Field2": ['abc', 'def', 'hgi']},
                        optional:
                         ['Field1', 'Field2']
                         ["rowOne", "rowTwo", "thirdRow"])

A DataFrame is basically a table with rows and columns.

Columns are named, rows are numbered (but can be named) and can be easily selected and calculated upon. Internally, columns are stored as 1d numpy arrays. If you set row names, they’re converted into a dictionary for fast access. There is a rich subselection/slicing API, see help(DataFrame.get_item) (it also works for setting values). Please note that any slice get’s you another DataFrame, to access individual entries use get_row(), get_column(), get_value().

DataFrames also understand basic arithmetic and you can either add (multiply,…) a constant value, or another DataFrame of the same size / with the same column names, like this:

#multiply every value in ColumnA that is smaller than 5 by 6.
my_df[my_df[:,'ColumnA'] < 5, 'ColumnA'] *= 6

#you always need to specify both row and column selectors, use : to mean everything
my_df[:, 'ColumnB'] = my_df[:,'ColumnA'] + my_df[:, 'ColumnC']

#let's take every row that starts with Shu in ColumnA and replace it with a new list (comprehension)
select = my_df.where(lambda row: row['ColumnA'].startswith('Shu'))
my_df[select, 'ColumnA'] = [row['ColumnA'].replace('Shu', 'Sha') for row in my_df[select,:].iter_rows()]

Dataframes talk directly to R via rpy2 (rpy2 is not a prerequiste for the library!)

 

(2) pandas http://pandas.pydata.org/

Library Highlights

  • A fast and efficient DataFrame object for data manipulation with integrated indexing;
  • Tools for reading and writing data between in-memory data structures and different formats: CSV and text files, Microsoft Excel, SQL databases, and the fast HDF5 format;
  • Intelligent data alignment and integrated handling of missing data: gain automatic label-based alignment in computations and easily manipulate messy data into an orderly form;
  • Flexible reshaping and pivoting of data sets;
  • Intelligent label-based slicing, fancy indexing, and subsetting of large data sets;
  • Columns can be inserted and deleted from data structures for size mutability;
  • Aggregating or transforming data with a powerful group by engine allowing split-apply-combine operations on data sets;
  • High performance merging and joining of data sets;
  • Hierarchical axis indexing provides an intuitive way of working with high-dimensional data in a lower-dimensional data structure;
  • Time series-functionality: date range generation and frequency conversion, moving window statistics, moving window linear regressions, date shifting and lagging. Even create domain-specific time offsets and join time series without losing data;
  • The library has been ruthlessly optimized for performance, with critical code paths compiled to C;
  • Python with pandas is in use in a wide variety of academic and commercial domains, including Finance, Neuroscience, Economics, Statistics, Advertising, Web Analytics, and more.

Why not R?

First of all, we love open source R! It is the most widely-used open source environment for statistical modeling and graphics, and it provided some early inspiration for pandas features. R users will be pleased to find this library adopts some of the best concepts of R, like the foundational DataFrame (one user familiar with R has described pandas as “R data.frame on steroids”). But pandas also seeks to solve some frustrations common to R users:

  • R has barebones data alignment and indexing functionality, leaving much work to the user. pandas makes it easy and intuitive to work with messy, irregularly indexed data, like time series data. pandas also provides rich tools, like hierarchical indexing, not found in R;
  • R is not well-suited to general purpose programming and system development. pandas enables you to do large-scale data processing seamlessly when developing your production applications;
  • Hybrid systems connecting R to a low-productivity systems language like Java, C++, or C# suffer from significantly reduced agility and maintainability, and you’re still stuck developing the system components in a low-productivity language;
  • The “copyleft” GPL license of R can create concerns for commercial software vendors who want to distribute R with their software under another license. Python and pandas use more permissive licenses.

(3) datamatrix http://pypi.python.org/pypi/datamatrix/0.8

datamatrix 0.8

A Pythonic implementation of R’s data.frame structure.

Latest Version: 0.9

This module allows access to comma- or other delimiter separated files as if they were tables, using a dictionary-like syntax. DataMatrix objects can be manipulated, rows and columns added and removed, or even transposed

—————————————————————–

Modeling in Python

Continue reading

Using Rapid Miner and R for Sports Analytics #rstats

Rapid Miner has been one of the oldest open source analytics software, long long before open source or even analytics was considered a fashion buzzword. The Rapid Miner software has been a pioneer in many areas (like establishing a marketplace for Rapid Miner Extensions) and the Rapid Miner -R extension was one of the most promising enablers of using R in an enterprise setting.
The following interview was taken with a manager of analytics for a sports organization. The sports organization considers analytics as a strategic differentiator , hence the name is confidential. No part of the interview has been edited or manipulated.

Ajay- Why did you choose Rapid Miner and R? What were the other software alternatives you considered and discarded?

Analyst- We considered most of the other major players in statistics/data mining or enterprise BI.  However, we found that the value proposition for an open source solution was too compelling to justify the premium pricing that the commercial solutions would have required.  The widespread adoption of R and the variety of packages and algorithms available for it, made it an easy choice.  We liked RapidMiner as a way to design structured, repeatable processes, and the ability to optimize learner parameters in a systematic way.  It also handled large data sets better than R on 32-bit Windows did.  The GUI, particularly when 5.0 was released, made it more usable than R for analysts who weren’t experienced programmers.

Ajay- What analytics do you do think Rapid Miner and R are best suited for?

 Analyst- We use RM+R mainly for sports analysis so far, rather than for more traditional business applications.  It has been quite suitable for that, and I can easily see how it would be used for other types of applications.

 Ajay- Any experiences as an enterprise customer? How was the installation process? How good is the enterprise level support?

Analyst- Rapid-I has been one of the most responsive tech companies I’ve dealt with, either in my current role or with previous employers.  They are small enough to be able to respond quickly to requests, and in more than one case, have fixed a problem, or added a small feature we needed within a matter of days.  In other cases, we have contracted with them to add larger pieces of specific functionality we needed at reasonable consulting rates.  Those features are added to the mainline product, and become fully supported through regular channels.  The longer consulting projects have typically had a turnaround of just a few weeks.

 Ajay- What challenges if any did you face in executing a pure open source analytics bundle ?

Analyst- As Rapid-I is a smaller company based in Europe, the availability of training and consulting in the USA isn’t as extensive as for the major enterprise software players, and the time zone differences sometimes slow down the communications cycle.  There were times where we were the first customer to attempt a specific integration point in our technical environment, and with no prior experiences to fall back on, we had to work with Rapid-I to figure out how to do it.  Compared to the what traditional software vendors provide, both R and RM tend to have sparse, terse, occasionally incomplete documentation.  The situation is getting better, but still lags behind what the traditional enterprise software vendors provide.

 Ajay- What are the things you can do in R ,and what are the things you prefer to do in Rapid Miner (comparison for technical synergies)

Analyst- Our experience has been that RM is superior to R at writing and maintaining structured processes, better at handling larger amounts of data, and more flexible at fine-tuning model parameters automatically.  The biggest limitation we’ve had with RM compared to R is that R has a larger library of user-contributed packages for additional data mining algorithms.  Sometimes we opted to use R because RM hadn’t yet implemented a specific algorithm.  The introduction the R extension has allowed us to combine the strengths of both tools in a very logical and productive way.

In particular, extending RapidMiner with R helped address RM’s weakness in the breadth of algorithms, because it brings the entire R ecosystem into RM (similar to how Rapid-I implemented much of the Weka library early on in RM’s development).  Further, because the R user community releases packages that implement new techniques faster than the enterprise vendors can, this helps turn a potential weakness into a potential strength.  However, R packages tend to be of varying quality, and are more prone to go stale due to lack of support/bug fixes.  This depends heavily on the package’s maintainer and its prevalence of use in the R community.  So when RapidMiner has a learner with a native implementation, it’s usually better to use it than the R equivalent.