Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 4 years ago.
Improve this question
As a student of computational linguistics, I frequently do machine learning experiments where I have to prepare training data from all kinds of different resources like raw or annotated text corpora or syntactic tree banks. For every new task and every new experiment I write programs (normally in Python and sometimes Java) to extract the features and values I need and transform the data from one format to the other. This usually results in a very large number of very large files and a very large number of small programs which process them in order to get the input for some machine learning framework (like the arff files for Weka).
One needs to be extremely well organised to deal with that and program with great care not to miss any important peculiarities, exceptions or errors in the tons of data. Many principles of good software design like design patterns or refactoring paradigms are no big use for these tasks because things like security, maintainability or sustainability are of no real importance - once the program successfully processed the data one doesn't need it any longer. This has gone so far that I even stopped bothering about using classes or functions at all in my Python code and program in a simple procedural way. The next experiment will require different data sets with unique characteristics and in a different format so that their preparation will likely have to be programmed from scratch anyway. My experience so far is that it's not unusual to spend 80-90% of a project's time on the task of preparing training data. Hours and days go by only on thinking about how to get from one data format to another. At times, this can become quite frustrating.
Well, you probably guessed that I'm exaggerating a bit, on purpose even, but I'm positive you understand what I'm trying to say. My question, actually, is this:
Are there any general frameworks, architectures, best practices for approaching these tasks? How much of the code I write can I expect to be reusable given optimal design?
I find myself mostly using the textutils from GNU coreutils and flex for corpus preparation, chaining things together in simple scripts, at least when the preparations i need to make are simple enough for regular expressions and trivial filtering etc.
It is still possible to make things reusable, the general rules also apply here. If you are programming with no regard to best practices and the like and just program procedurally there is IMHO really no wonder that you have to do everything from scratch when starting a new project.
Even though the format requirements will vary a lot there is still many common tasks, ie. tag-stripping, tag-translation, selection, tabulation, some trivial data harvesting such as number of tokens, sentences and the like. Programming these tasks aming for high reusability will pay off, even though it takes longer at first.
I am not aware of any such frameworks--doesn't mean they aren't out there. I prefer to use my own which is just a collection of code snippets i've refined/tweaked/borrowed over time and that i can chain together in various configurations depending on the problem. If you already know python, then i strongly recommend handling all of your data prep in NumPy--as you know, ML data sets tends to be large--thousands of row vectors packed with floats. NumPy is brilliant for that sort of thing. Additionally, I might suggest that for preparing training data for ML, there are a couple of tasks that arise in nearly every such effort and that don't vary a whole lot from one problem to the next. I've give you snippets for these below.
normalization (scaling & mean-centering your data to avoid overweighting. As i'm sure you know, you can scale -1 to 1 or 0 to 1. I usually chose the latter so that i can take advantage of sparsity patterns. In python, using the NumPy library:
import numpy as NP
data = NP.linspace( 1, 12, 12).reshape(4, 3)
data_norm = NP.apply_along_axis( lambda x : (x - float(x.min())) / x.max(),
0, data )
cross-validation (here's i've set the default argument at '5', so test set is 5%, training set, 95%--putting this in a function makes k-fold much simpler)
def divide_data(data, testset_size=5) :
max_ndx_val = data.shape[0] -1
ndx2 = NP.random.random_integers(0, max_ndx_val, testset_size)
TE = data_rows[ndx2]
TR = NP.delete(data, ndx2, axis=0)
return TR, TE
Lastly, here's an excellent case study (IMHO), both clear and complete, showing literally the entire process from collection of the raw data through input to the ML algorithm (a MLP in this case). They also provide their code.
Related
Is it necessary to repeat similar template data... Like the meaning and context is the same, but the smaller details vary. If I remove these redundancies, the dataset is very small (size in hundreds) but if the data like these are included, it easily crosses thousands. Which is the right approach?
SAMPLE DATA
This is acutally not a question suited for stack overflow but I'll answer anyways:
You have to think about how the emails (or what ever your data this is) will look in real-life usage: Do you want to detect any kind of spam or just similiar to what your sample data shows? If the first is the case, your dataset is just not suited for this problem since there are not enough various data samples. When you think about it, every of the senteces are exactly the same because the company name isn't really valueable information and will probably not be learned as a feature by your RNN. So the information is almost the same. And since every input sample will run through the network multiple times (once each epoch) it doesnt really help having almost the same sample multiple times.
So you shouldnt have one kind of almost identical data samples dominating your dataset.
But as I said: When you primarily want to filter out "Dear customer, we wish you a ..." you can try it with this dataset but you wouldnt really need an RNN to detect that. If you want to detect all kind of spam, you should search for a new dataset since ~100 unique samples are not enough. I hope that was helpful!
[I'm approaching this as an outsider to machine learning. It just seems like a classification problem which I should be able to solve with fairly good accuracy with Machine Larning.]
Training Dataset:
I have millions of URLs, each tagged with a particular category. There are limited number of categories (50-100).
Now given a fresh URL, I want to categorize it into one of those categories. The category can be determined from the URL using conventional methods, but would require a huge unmanageable mess of pattern matching.
So I want to build a box where INPUT is URL, OUTPUT is Category. How do I build this box driven by ML?
As much as I would love to understand the basic fundamentals of how this would work out mathematically, right now much much more focussed on getting it done, so a conceptual understanding of the systems and processes involved is what I'm looking to get. I suppose machine learning is at a point where you can approach reasonably straight forward problems in that manner.
If you feel I'm wrong and I need to understand the foundations deeply in order to get value out of ML, do let me know.
I'm building this inside an AWS ecosystem so I'm open to using Amazon ML if it makes things quicker and simpler.
I suppose machine learning is at a point where you can approach reasonably straight forward problems in that manner.
It is not. Building an effective ML solution requires both an understanding of problem scope/constraints (in your case, new categories over time? Runtime requirements? Execution frequency? Latency requirements? Cost of errors? and more!). These constraints will then impact what types of feature engineering / processing you may look at, and what types of models you will look at. Your particular problem may also have issues with non I.I.D. data, which is an assumption of most ML methods. This would impact how you evaluate the accuracy of your model.
If you want to learn enough ML to do this problem, you might want to start looking at work done in Malicious URL classification. An example of which can be found here. While you could "hack" your way to something without learning more about ML, I would not personally trust any solution built in that manner.
If you feel I'm wrong and I need to understand the foundations deeply in order to get value out of ML, do let me know.
Okay, I'll bite.
There are really two schools of thought currently related to prediction: "machine learners" versus statisticians. The former group focuses almost entirely on practical and applied prediction, using techniques like k-fold cross-validation, bagging, etc., while the latter group is focused more on statistical theory and research methods. You seem to fall into the machine-learning camp, which is fine, but then you say this:
As much as I would love to understand the basic fundamentals of how this would work out mathematically, right now much much more focussed on getting it done, so a conceptual understanding of the systems and processes involved is what I'm looking to get.
While a "conceptual understanding of the systems and processes involved" is a prerequisite for doing advanced analytics, it isn't sufficient if you're the one conducting the analysis (it would be sufficient for a manager, who's not as close to the modeling).
With just a general idea of what's going on, say, in a logistic regression model, you would likely throw all statistical assumptions (which are important) to the wind. Do you know whether certain features or groups shouldn't be included because there aren't enough observations in that group for the test statistic to be valid? What can happen to your predictions and hypotheses when you have high variance-inflation factors?
These are important considerations when doing statistics, and oftentimes people see how easy it is to do from sklearn.svm import SVC or somthing like that and run wild. That's how you get caught with your pants around your ankles.
How do I build this box driven by ML?
You don't seem to have even a rudimentary understanding of how to approach machine/statistical learning problems. I would highly recommend that you take an "Introduction to Statistical Learning"- or "Intro to Regression Modeling"-type course in order to think about how you translate the URLs you have into meaningful features that have significant power predicting URL class. Think about how you can decompose a URL into individual pieces that might give some information as to which class a certain URL pertains. If you're classifying espn.com domains by sport, it'd be pretty important to parse nba out of http://www.espn.com/nba/team/roster/_/name/cle, don't you think?
Good luck with your project.
Edit:
To nudge you along, though: every ML problem boils down to some function mapping input to output. Your outputs are URL classes. Your inputs are URLs. However, machines only understand numbers, right? URLs aren't numbers (AFAIK). So you'll need to find a way to translate information contained in the URLs to what we call "features" or "variables." One place to start, there, would be one-hot encoding different parts of each URL. Think of why I mentioned the ESPN example above, and why I extracted info like nba from the URL. I did that because, if I'm trying to predict to which sport a given URL pertains, nba is a dead giveaway (i.e. it would very likely be highly predictive of sport).
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 3 years ago.
Improve this question
Can somebody explain me the main pros and cons of the most known datamining open-source tools?
Everywhere I read that RapidMiner, Weka, Orange, KNIME are the best ones.
look at this blog post
Can somebody do a fast technical comparison in a small bullet list.
My needs are the following:
It should support classification algorithms (Naive Bayes, SVM, C4.5,
kNN).
It should be easy to implement in Java.
It should have understandable documentation.
It should have reference production projects or use cases working on in.
some additional benchmark comparison if possible.
Thanks!
I would like to say firstly there are pro's and cons for each of them on your list however I would suggest out of your list weka from my personal experience it is incredibly simple to implement in your own java application using the weka jar file and has its own self contained tools for data mining.
Rapid miner seems to be a commercial solution offering an end to end solution however the most notable number of examples of external implementations of solutions for rapid miner are usually in python and r script not java.
Orange offers tools that seem to be targeted primarily at people with possibly less need for custom implementations into their own software but a far easier time with user itneraction, its written in python and source is available, user addons are supported.
Knime is another commercial platform offering end to end solutions for data mining and analysis providing all the tools required, this one has various good reviews around the internet but i havent used it enough to advise you or anyone on the pro's or cons of it.
See here for knime vs weka
Best data mining tools
As i said weka is my personal favorite as a software developer but im sure other people have varying reasons and opinions on why to choose one over the other. Hope you find the right solution for you.
Also per your requirements weka supports the following:
Naivebayes
SVM
C4.5
KNN
I have tried Orange and Weka with a 15K records database and found problems with the memory management in Weka, it needed more than 16Gb of RAM while Orange could've managed the database without using that much. Once Weka reaches the maximum amount of memory, it crashes, even if you set more memory in the ini file telling Java virtual machine to use more.
I recently evaluated many open source projects, comparing and contrasted them with regards to the decision tree machine learning algorithm. Weka and KNIME were included in that evaluation. I covered the differences in algorithm, UX, accuracy, and model inspection. You might chose one or the other depending on what features you value most.
I have had positive experience with RapidMiner:
a large set of machine learning algorithms
machine learning tools - feature selection, parameter grid search, data partitioning, cross validation, metrics
a large set of data manipulation algorithms - input, transformation, output
applicable to many domains - finance, web crawling and scraping, nlp, images (very basic)
extensible - one can send and receive data other technologies: R, python, groovy, shell
portable - can be run as a java process
developer friendly (to some extent, could use some improvements) - logging, debugging, breakpoints, macros
I would have liked to see something like RapidMiner in terms of user experience, but with the underlying engine based on python technologies: pandas, scikit-learn, spacy etc. Preferably, something that would allow moving back and forth from GUI to code.
I asked myself the question whether most people normally code the machine learning algorithms themselves or whether they are likely to use existing solutions like Weka or R packages.
Of course it depends on the problem - but let's say that I want to use a common solution like a neural network. Is there still a reason to code it myself? To understand the mechanism better and adapt it? Or is the thought of standardized solutions more important?
This is not a good question for Stackoverflow. It's an opinion question, not a programming problem.
Nevertheless, here is my take:
It depends on what you want to do.
If you want to find which algorithm works best for your data problem at hand, try ELKI, Weka, R, Matlab, SciPy, whatever. Try out all the algorithms you can find, and spend even more time on preprocessing your data.
If you know which algorithm you need and need to get it into production, many of these tools will not perform good enough or be easy enough to integrate. Instead, check if you can find low level libraries such as libSVM that provide the functionality you need. If these don't exist, roll your own optimized code.
If you want to do research in this domain, you are best off with extending the existing tools. ELKI and Weka have APIs that you can plug into to provide extensions. R doesn't really have an API (CRAN it's a mess...) but people just dump their code somewhere and (hopefully) add a manual how to use it. Extending these frameworks can save you a lot of effort: you have comparison methods ready to use, and you can re-use a lot of their code. ELKI for example has a lot of index structures to accelerate algorithms. Most of the time, the index acceleration is much harder to write than the actual algorithm. So if you can reuse the existing indexes, this will make your algorithms much faster, too (and you will also benefit from future enhancements to these frameworks).
If you want to learn about existing algorithms you better implement them yourself. You'll be surprised how much more there is to optimizing some algorithms than what is taught in class. E.g. APRIORI. The basic idea is quite simple. But getting all the pruning details right, I say 1 out of 20 students gets these details. If you implement APRIORI, then benchmark it against a known good implementation and try to understand why yours is much slower, then you'll actually discover the subtle details to the algorithms. And don't be surprised to see a factor of 100 performance difference between ELKI, R, Weka etc. - it's can still be the same algorithm, just implemented more or less efficiently when it comes to actual data structures used, memory layout etc.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 2 years ago.
Improve this question
In the beginning, I would like to describe my current position and the goal that I would like to achieve.
I am a researcher dealing with machine learning. So far have gone through several theoretical courses covering machine learning algorithms and social network analysis and therefore have gained some theoretical concepts useful for implementing machine learning algorithms and feed in the real data.
On simple examples, the algorithms work well and the running time is acceptable whereas the big data represent a problem if trying to run algorithms on my PC. Regarding the software I have enough experience to implement whatever algorithm from articles or design my own using whatever language or IDE (so far have used Matlab, Java with Eclipse, .NET...) but so far haven't got much experience with setting-up infrastructure. I have started to learn about Hadoop, NoSQL databases, etc, but I am not sure what strategy would be the best taking into consideration the learning time constraints.
The final goal is to be able to set-up a working platform for analyzing big data with focusing on implementing my own machine learning algorithms and put all together into production, ready for solving useful question by processing big data.
As the main focus is on implementing machine learning algorithms I would like to ask whether there is any existing running platform, offering enough CPU resources to feed in large data, upload own algorithms and simply process the data without thinking about distributed processing.
Nevertheless, such a platform exists or not, I would like to gain a picture big enough to be able to work in a team that could put into production the whole system tailored upon the specific customer demands. For example, a retailer would like to analyze daily purchases so all the daily records have to be uploaded to some infrastructure, capable enough to process the data by using custom machine learning algorithms.
To put all the above into simple question: How to design a custom data mining solution for real-life problems with main focus on machine learning algorithms and put it into production, if possible, by using the existing infrastructure and if not, design distributed system (by using Hadoop or whatever framework).
I would be very thankful for any advice or suggestions about books or other helpful resources.
First of all, your question needs to define more clearly what you intend by Big Data.
Indeed, Big Data is a buzzword that may refer to various size of problems. I tend to define Big Data as the category of problems where the Data size or the Computation time is big enough for "the hardware abstractions to become broken", which means that a single commodity machine cannot perform the computations without intensive care of computations and memory.
The scale threshold beyond which data become Big Data is therefore unclear and is sensitive to your implementation. Is your algorithm bounded by Hard-Drive bandwidth ? Does it have to feet into memory ? Did you try to avoid unnecessary quadratic costs ? Did you make any effort to improve cache efficiency, etc.
From several years of experience in running medium large-scale machine learning challenge (on up to 250 hundreds commodity machine), I strongly believe that many problems that seem to require distributed infrastructure can actually be run on a single commodity machine if the problem is expressed correctly. For example, you are mentioning large scale data for retailers. I have been working on this exact subject for several years, and I often managed to make all the computations run on a single machine, provided a bit of optimisation. My company has been working on simple custom data format that allows one year of all the data from a very large retailer to be stored within 50GB, which means a single commodity hard-drive could hold 20 years of history. You can have a look for example at : https://github.com/Lokad/lokad-receiptstream
From my experience, it is worth spending time in trying to optimize algorithm and memory so that you could avoid to resort to distributed architecture. Indeed, distributed architectures come with a triple cost. First of all, the strong knowledge requirements. Secondly, it comes with a large complexity overhead in the code. Finally, distributed architectures come with a significant latency overhead (with the exception of local multi-threaded distribution).
From a practitioner point of view, being able to perform a given data mining or machine learning algorithm in 30 seconds is one the key factor to efficiency. I have noticed than when some computations, whether sequential or distributed, take 10 minutes, my focus and efficiency tend to drop quickly as it becomes much more complicated to iterate quickly and quickly test new ideas. The latency overhead introduced by many of the distributed frameworks is such that you will inevitably be in this low-efficiency scenario.
If the scale of the problem is such that even with strong effort you cannot perform it on a single machine, then I strongly suggest to resort to on-shelf distributed frameworks instead of building your own. One of the most well known framework is the MapReduce abstraction, available through Apache Hadoop. Hadoop can be run on 10 thousands nodes cluster, probably much more than you will ever need. If you do not own the hardware, you can "rent" the use of a Hadoop cluster, for example through Amazon MapReduce.
Unfortunately, the MapReduce abstraction is not suited to all Machine Learning computations.
As far as Machine Learning is concerned, MapReduce is a rigid framework and numerous cases have proved to be difficult or inefficient to adapt to this framework:
– The MapReduce framework is in itself related to functional programming. The
Map procedure is applied to each data chunk independently. Therefore, the
MapReduce framework is not suited to algorithms where the application of the
Map procedure to some data chunks need the results of the same procedure to
other data chunks as a prerequisite. In other words, the MapReduce framework
is not suited when the computations between the different pieces of data are
not independent and impose a specific chronology.
– MapReduce is designed to provide a single execution of the map and of the
reduce steps and does not directly provide iterative calls. It is therefore not
directly suited for the numerous machine-learning problems implying iterative
processing (Expectation-Maximisation (EM), Belief Propagation, etc.). The
implementation of these algorithms in a MapReduce framework means the
user has to engineer a solution that organizes results retrieval and scheduling
of the multiple iterations so that each map iteration is launched after the reduce
phase of the previous iteration is completed and so each map iteration is fed
with results provided by the reduce phase of the previous iteration.
– Most MapReduce implementations have been designed to address production needs and
robustness. As a result, the primary concern of the framework is to handle
hardware failures and to guarantee the computation results. The MapReduce efficiency
is therefore partly lowered by these reliability constraints. For example, the
serialization on hard-disks of computation results turns out to be rather costly
in some cases.
– MapReduce is not suited to asynchronous algorithms.
The questioning of the MapReduce framework has led to richer distributed frameworks where more control and freedom are left to the framework user, at the price of more complexity for this user. Among these frameworks, GraphLab and Dryad (both based on Direct Acyclic Graphs of computations) are well-known.
As a consequence, there is no "One size fits all" framework, such as there is no "One size fits all" data storage solution.
To start with Hadoop, you can have a look at the book Hadoop: The Definitive Guide by Tom White
If you are interested in how large-scale frameworks fit into Machine Learning requirements, you may be interested by the second chapter (in English) of my PhD, available here: http://tel.archives-ouvertes.fr/docs/00/74/47/68/ANNEX/texfiles/PhD%20Main/PhD.pdf
If you provide more insight about the specific challenge you want to deal with (type of algorithm, size of the data, time and money constraints, etc.), we probably could provide you a more specific answer.
edit : another reference that could prove to be of interest : Scaling-up Machine Learning
I had to implement a couple of Data Mining algorithms to work with BigData too, and I ended up using Hadoop.
I don't know if you are familiar to Mahout (http://mahout.apache.org/), which already has several algorithms ready to use with Hadoop.
Nevertheless, if you want to implement your own Algorithm, you can still adapt it to Hadoop's MapReduce paradigm and get good results. This is an excellent book on how to adapt Artificial Intelligence algorithms to MapReduce:
Mining of Massive Datasets - http://infolab.stanford.edu/~ullman/mmds.html
This seems to be an old question. However given your usecase, the main frameworks focusing on Machine Learning in Big Data domain are Mahout, Spark (MLlib), H2O etc. However to run Machine Learning algorithms on Big Data you have to convert them to parallel programs based on Map Reduce paradigm. This is a nice article giving a brief introduction to major (not all) big Data frameworks:
http://www.codophile.com/big-data-frameworks-every-programmer-should-know/
I hope this will help.