Given a word I want to get the list of most frequent predecessors and successors of the word in English language.
I have developed a code that does bigram analysis on any corpus ( I have used Enron email corpus) and can predict the most frequent next possible word but I want some other solution because
a) I want to check the working / accuracy of my prediction
b) Corpus or dataset based solutions fail for an unseen word
For example, given the word "excellent" I want to get the words that are most likely to come before excellent and after excellent
My question is whether any particular service or api exists for the purpose?
Any solution to this problem is bound to be a corpus-based method; you just need a bigger corpus. I'm not aware of any web service or library that is does this for you, but there are ways to obtain bigger corpora:
Google has published a huge corpus of n-grams collected from the English part of the web. It's available via the Linguistic Data Consortium (LDC), but I believe you must be an LDC member to obtain it. (Many universities are.)
If you're not an LDC member, try downloading a Wikipedia database dump (get enwiki) and training your predictor on that.
If you happen to be using Python, check out the nice set of corpora (and tools) delivered with NLTK.
As for the unseen words problem, there are ways to tackle it, e.g. by replacing all words that occur less often than some threshold by a special token like <unseen> prior to training. That will make your evaluation a bit harder.
You have got to give some more instances or context of "unseen" word so that the algorithm can make some inference.
One indirect way can be reading rest of the words in the sentences.. and looking into a dictionary for the words where those words are encountered.
In general, you cant expect the algorithm to learn and understand the inference in the first time. Think about yourself.. If you were given a new word.. how well can you make out its meaning (probably by looking into how it has been used in the sentence and how well your understanding is) but then you make an educated guess and over the period of time you understand the meaning.
I just re-read the original question and I realize the answers, mine included got off base. I think the original person just wanted to solve a simple programming problem, not look for datasets.
If you list all distinct word-pairs and count them, then you can answer your question with simple math on that list.
Of course you have to do a lot of processing to generate the list. While it's true that if the total number of distinct words is as much a 30,000 then there are a billion possible pairs, I doubt that in practice there are that many. So you can probably make a program with a huge hash table in memory (or on disk) and just count them all. If you don't need the insignificant pairs you could write a program that flushes out the less important ones periodically while scanning. Also you can segment the word list and generate pairs of a hundred words verses the rest, then the next hundred and so on, and calculate in passes.
My original answer is here I'm leaving it because it's my own related question:
I'm interested in something similar (I'm writing a entry system that suggest word completions and punctuation and I would like it to be multilingual).
I found a download page for google's ngram files, but they're not that good, they're full of scanning errors. 'i's become '1's, words run together etc. Hopefully Google has improved their scanning technology since then.
The just-download-wikipedia-unpack=it-and-strip-the-xml idea is a bust for me, I don't have a fast computer (heh, I have a choice between an atom netbook here and an android device). Imagine how long it would take me to unpack a 3 gigabytes of bz2 file becoming what? 100 of xml, then process it with beautiful soup and filters that he admits crash part way through each file and need to be restarted.
For your purpose (previous and following words) you could create a dictionary of real words and filter the ngram lists to exclude the mis-scanned words. One might hope that the scanning was good enough that you could exclude misscans by only taking the most popular words... But I saw some signs of constant mistakes.
The ngram datasets are here by the way http://books.google.com/ngrams/datasets
This site may have what you want http://www.wordfrequency.info/
Related
Edit: some people flagged this question as a potential duplicate of this other one. While I agree that knowing how the birthday paradox applies to hashing functions, the 2 questions (and respective answers) address 2 different, albeit related, subjects.
The other question is asking "what are the odds of collision", whereas this question main focus is "how can I make sure that collision never happens".
I have a data lake stored in S3 where each day an ETL script dumps additional data from the day before.
Due to how the pipeline is built, it is possible for a very inconsiderate user that has admin access to produce duplicates in said data lake by manually interacting with the dump files coming from our OLTP database, and triggering the ETL script when it's not supposed to.
I thought that a good idea to prevent data duplication was to insert a form of security measure in my ETL script:
Produce a hash for each entry.
Store said hashes somewhere else (like a dynamodb table).
Whenever new data comes in, hash that as well and compare it with the already existing hashes.
If any of new hash is in the existing hashes, reject the associated entry entirely.
However, I know very little about hashing and I was reading that, although unlikely, 2 different sources can produce the same hash.
I understand it's really hard for it to happen in this situation, but I was wondering if there is a way to be 100% sure about it.
Any idea is much appreciated.
Long answer: what you want to study and explore is called "perfect hashing" (ie hashing guaranteed not to have collisions. https://en.wikipedia.org/wiki/Perfect_hash_function
Short answer: A cryptographic collision resistant algorithm like sha-1 is probably safe to use for all but the largest (PBs a day) datasets and even then its probably all right. Git uses sha-1 internally and code repositories probably deal with the most files on the planet and rarely have collisions.
See for details: https://ericsink.com/vcbe/html/cryptographic_hashes.html#:~:text=Git%20uses%20hashes%20in%20two,computed%20when%20it%20was%20stored.
Medium answer: this is actually a pretty hard problem overall and a frequent area of study for computer science and a lot depends on your particular use case and the context you're operating in. Cuckoo hashing, collision resistant algorithms, and hashing in general are probably all good terms to research. There's also a lot of art and science behind space (memory) and time (computer power needed) when picking these methods. A good rule of thumb is that perfect hashing will generally take up more space and time than a collision resistant cryptographic hash like sha-1.
To find a synchronizing word I have always just used trial and error, which for small DFAs is fine but not so useful on larger DFAs. What I want to know, however, is if there exists an algorithm for determining a synchronizing word or if there is a way of being able to tell that one does not exist. (Rather than just saying "I can't find one, therefore one can not exist" which is by no means a proof).
I have had a look around on google and so far just came across methods for determining what the upper and lower bounds for a length of a synchronizing word would be based on the number of states, however this is not helpful to me.
The existence of upper bounds on the length of a synchronizing word immediately implies the existence of a (very slow) algorithm for finding one: just list all strings of length less than the upper bound and test whether each is a synchronizing word. If any of them are, then the synchronizing word exists, and if none of them are, there's no synchronizing word. This is exponentially slow, though, so it's not advisable on large DFAs.
David Eppstein designed a polynomial-time algorithm for finding synchronizing words in DFAs, though I'm not very familiar with this algorithm.
Hope this helps!
I'm maintaining an archive of the heavily redacted documents coming out of the Foreign Intelligence Surveillance Court.
They come with big sections of text that look like this:
And when the OCR tries to work with this, you get text like:
production of this data on a daily basis for a period of 90 days. The sole purpose of this
production is to obtain foreign intelligence information in support of
individual authorized investigations to protect against international terrorism and
So in the OCRed version, where there are blacked out spots, there are just missing words. Sometimes, the missing words create a grammatically correct sentence with a different/weird meaning (like above). Other times, the resulting sentences make no sense, but either way it's a problem. It would be much better if the OCR engine could return X's for these spots or Unicode squares like ▮▮▮▮ instead.
The result I'd like is something like:
production of this data on a daily basis for a period of 90 days. The sole purpose of this
production is to obtain foreign intelligence information in support of XXXXXXXXXXX
individual authorized investigations to protect against international terrorism and
My question is how to go about getting these X's. Is there a way to analyze the images to identify the black spots? Is there a way to replace them with X's or some better unicode character? I'm open to any ideas to make this look right, but image editing is not a strong suit for me nor is hacking deep within the OCR engine.
You may want to train Tesseract for those long blobs. Depending on the length of the blob, you would assign a different number of 'X' characters. Read TrainingTesseract3 for training process.
Folks,
I have a bunch of documents (approx 200k) that have a title and abstract. There is other meta data available for each document for example category - (only one of cooking, health, exercise etc), genre - (only one of humour, action, anger) etc. The meta data is well structured and all this is available in a MySql DB.
I need to show to our user related documents while she is reading one of these document on our site. I need to provide the product managers weight-ages for title, abstract and meta data to experiment with this service.
I am planning to run clustering on top of this data, but am hampered by the fact that all Mahout Clustering example use either DenseVectors formulated on top of numbers, or Lucene based text vectorization.
The examples are either numeric data only or text data only. Has any one solved this kind of a problem before. I have been reading Mahout in Action book and the Mahout Wiki, without much success.
I can do this from the first principles - extract all titles and abstracts in to a DB, calculate TFIDF & LLR, treat each word as a dimension and go about this experiment with a lot of code writing. That seems like a longish way to the solution.
That in a nutshell is where I am trapped - am I doomed to the first principles or there exist a tool / methodology that I somehow missed. I would love to hear from folks out there who have solved similar problem.
Thanks in advance
You have a text similarity problem here and I think you're thinking about it correctly. Just follow any example concerning text. Is it really a lot of code? Once you count the words in the docs you're mostly done. Then feed it into whatever clusterer you want. The term extractions is not something you do with Mahout, though there are certainly libraries and tools that are good at it.
I'm actually working on something similar, but without the need of distinciton between numeric and text fields.
I have decided to go with the semanticvectors package which does all the part about tfidf, the semantic space vectors building, and the similarity search. It uses a lucene index.
Please note that you can also use the s-space package if semanticvectors doesn't suit you (if you go down that road of course).
The only caveat I'm facing with this approach is that the indexing part can't be iterative. I have to index everything every time a new document is added, or an old document is modified. People using semanticvectors say they have very good indexing times. But I don't know how large their corpora are. I'm going to test these issues with the wikipedia dump to see how fast it can be.
I'm working on a project with designing a core data system for searching and cataloguing images and documents. One of the objects in my data model is a 'key word' object. Every time I add a new key word I first want to first run though all of the existing keywords to make sure it doesn't already exist in the current context.
I've read in posts here and in a lot of my reading that doing string comparisons is a far more expensive processing than some other comparison operations. Since I could easily end up having to check many thousands of words before a new addition I'm wondering if it would be worth using some method that would represent the key word strings numerically for the purpose of this process. Possibly breaking down each character in the string into a number formed from the UTF code for each character and then storing that in an ID property for each key word.
I was wondering if anyone else thought any benefit might come from this approach or if anyone else had any better ideas.
What you might useful is a suitable hash function to convert your text strings into (probably) unique numbers. (You might still have to check for collision effects.)
Comparing intrinsic numbers in C code is a much faster for several reasons. It avoids the Objective C runtime dispatch overhead. It requires accessing less total memory. And the executable code for each comparison is usually just an instruction or 3, rather than a loop with incrementers and several decision points.