Resource that provides number of documents where the term is covered - nlp

I am looking for resources that provides the number of documents a term is covered in. For example, there is about 25 billion documents that contains the term "the" in the indexed internet.

I don't know of any document frequency lists for large corpora such as the web, but there are some term frequency lists available. For example, there are the frequency lists from the web corpora compiled by the Web-As-Corpus Kool Yinitiative, which include the 2-billion ukWaC English web corpus. Alternatively, there are the n-grams from the Google Books Corpus.
It has been shown that such term frequency counts can be used to reliably approximate document frequency counts.

Here is a little more treatable frequencies.
Also take a look at this site - it contains a lot of info about existing corpora and words/ngrams lists. Unfortunately, most resources are paid, but not n-grams (for n > 1), so if you're going to process multiword terms, it can help.

Related

Word frequency database with senses

I am looking for a downloadable database of word frequencies/probabilities including their senses. Ideally, if it was mapped with WordNet.
In the list, some words would be listed multiple times if they have multiple senses e.g. frequency for 'bank' as an institution would be greater than river 'bank'.
Other datasets showing frequencies by word/part of speech would be helpful too.
Thanks for reading this.
N gram Frequencies are available in Google Ngram data. Although this does not answer the wordnet or the "senses" part of the question, it is a good start.
Use this package for experimenting the same

Applied NLP: how to score a document against a lexicon of multi-word terms?

This is probably a fairly basic NLP question but I have the following task at hand: I have a collection of text documents that I need to score against an (English) lexicon of terms that could be 1-, 2-, 3- etc N-word long. N is bounded by some "reasonable" number but the distribution of various terms in the dictionary for various values of n = 1, ..., N might be fairly uniform. This lexicon can, for example, contain a list of devices of certain type and I want to see if a given document is likely about any of these devices. So I would want to score a document high(er) if it has one or more occurrences of any of the lexicon entries.
What is a standard NLP technique to do the scoring while accounting for various forms of the words that may appear in the lexicon? What sort of preprocessing would be required for both the input documents and the lexicon to be able to perform the scoring? What sort of open-source tools exist for both the preprocessing and the scoring?
I studied LSI and topic modeling almost a year ago, so what I say should be taken as merely a pointer to give you a general idea of where to look.
There are many different ways to do this with varying degrees of success. This is a hard problem in the realm of information retrieval. You can search for topic modeling to learn about different options and state of the art.
You definitely need some preprocessing and normalization if the words could appear in different forms. How about NLTK and one of its stemmers:
>>> from nltk.stem.lancaster import LancasterStemmer
>>> st = LancasterStemmer()
>>> st.stem('applied')
'apply'
>>> st.stem('applies')
'apply'
You have a lexicon of terms that I am going to call terms and also a bunch of documents. I am going to explore a very basic technique to rank documents with regards to the terms. There are a gazillion more sophisticated ways you can read about, but I think this might be enough if you are not looking for something too sophisticated and rigorous.
This is called a vector space IR model. Terms and documents are both converted to vectors in a k-dimensional space. For that we have to construct a term-by-document matrix. This is a sample matrix in which the numbers represent frequencies of the terms in documents:
So far we have a 3x4 matrix using which each document can be expressed by a 3-dimensional array (each column). But as the number of terms increase, these arrays become too large and increasingly sparse. Also, there are many words such as I or and that occur in most of the documents without adding much semantic content. So you might want to disregard these types of words. For the problem of largeness and sparseness, you can use a mathematical technique called SVD that scales down the matrix while preserving most of the information it contains.
Also, the numbers we used on the above chart were raw counts. Another technique would be to use Boolean values: 1 for presence and 0 zero for lack of a term in a document. But these assume that words have equal semantic weights. In reality, rarer words have more weight than common ones. So, a good way to edit the initial matrix would be to use ranking functions like tf-id to assign relative weights to each term. If by now we have applied SVD to our weighted term-by-document matrix, we can construct the k-dimensional query vectors, which are simply an array of the term weights. If our query contained multiple instances of the same term, the product of the frequency and the term weight would have been used.
What we need to do from there is somewhat straightforward. We compare the query vectors with document vectors by analyzing their cosine similarities and that would be the basis for the ranking of the documents relative to the queries.

tf-idf using data on unigram frequency from Google

I'm trying to identify important terms in a set of government documents. Generating the term frequencies is no problem.
For document frequency, I was hoping to use the handy Python scripts and accompanying data that Peter Norvig posted for his chapter in "Beautiful Data", which include the frequencies of unigrams in a huge corpus of data from the Web.
My understanding of tf-idf, however, is that "document frequency" refers to the number of documents containing a term, not the number of total words that are this term, which is what we get from the Norvig script. Can I still use this data for a crude tf-idf operation?
Here's some sample data:
word tf global frequency
china 1684 0.000121447
the 352385 0.022573582
economy 6602 0.0000451130774123
and 160794 0.012681757
iran 2779 0.0000231482902018
romney 1159 0.000000678497795593
Simply dividing tf by gf gives "the" a higher score than "economy," which can't be right. Is there some basic math I'm missing, perhaps?
As I understand, Global Frequency is equal "inverse total term frequency" mentioned here Robertson. From this Robertson's paper:
One possible way to get away from this problem would be to make a fairly radical re-
placement for IDF (that is, radical in principle, although it may be not so radical
in terms of its practical effects). ....
the probability from the event space of documents to the event space of term positions
in the concatenated text of all the documents in the collection.
Then we have a new measure, called here
inverse total term frequency:
...
On the whole, experiments with inverse total term frequency weights have tended to show
that they are not as effective as IDF weights
According to this text, you can use inverse global frequency as IDF term, albeit more crude than standard one.
Also you are missing stop words removal. Words such as the are used in almost all documents, therefore they do not give any information. Before tf-idf , you should remove such stop words.

Financial news headers classification to positive/negative classes

I'm doing a small research project where I should try to split financial news articles headers to positive and negative classes.For classification I'm using SVM approach.The main problem which I see now it that not a lot of features can be produced for ML. News articles contains a lot of Named Entities and other "garbage" elements (from my point of view of course).
Could you please suggest ML features which can be used for ML training? Current results are: precision =0.6, recall=0.8
Thanks
The task is not trivial at all.
The straightforward approach would be to find or create a training set. That is a set of headers with positive news and a set of headers with negative news.
You turn the training set to a TF/IDF representation and then you train a Linear SVM to separate the two classes. Depending on the quality and size of your training set you can achieve something decent - not sure for 0.7 break even point.
Then, to get better results you need to go for NLP approaches. Try use a part-of-speech tagger to identify adjectives (trivial), and then score them using some sentiment DB like SentiWordNet.
There is an excellent overview on Sentiment Analysis by Bo Pang and Lillian Lee you should read:
How about these features?
Length of article header in words
Average word length
Number of words in a dictionary of "bad" words, e.g. dictionary = {terrible, horrible, downturn, bankruptcy, ...}. You may have to generate this dictionary yourself.
Ratio of words in that dictionary to total words in sentence
Similar to 3, but number of words in a "good" dictionary of words, e.g. dictionary = {boon, booming, employment, ...}
Similar to 5, but use the "good"-word dictionary
Time of the article's publication
Date of the article's publication
The medium through which it was published (you'll have to do some subjective classification)
A count of certain punctuation marks, such as the exclamation point
If you're allowed access to the actual article, you could use surface features from the actual article, such as its total length and perhaps even the number of responses or the level of opposition to that article. You could also look at many other dictionaries online such as Ogden's 850 basic english dictionary, and see if bad/good articles would be likely to extract many words from those. I agree that it seems difficult to come up with a long list (e.g. 100 features) of useful features for this purpose.
iliasfl is right, this is not a straightforward task.
I would use a bag of words approach but use a POS tagger first to tag each word in the headline. Then you could remove all of the named entities - which as you rightly point out don't affect the sentiment. Other words should appear frequently enough (if your dataset is big enough) to cancel themselves out from being polarised as either positive or negative.
One step further along, if you still aren't close could be to only select the adjectives and verbs from the tagged data as they are the words that tend to convey the emotion or mood.
I wouldn't be too disheartened in your precision and recall figures though, an F number of 0.8 and above is actually quite good.

tf-idf: Does using it help to weigh documents that share the terms higher than a document that doesnt?

I'm working on a customized search feature for a website. and I was curious if using only tf-idf to rank documents in my corpus would also help to weigh documents that have multiple search terms higher than documents with only one search term.
Example: Search = "poland spring water"
Theoretically, would the above query weigh (using traditional tf-idf) a document higher if the a document contained "poland" 100 times and "water" zero times. Or would it weigh a document heavier if it contained "poland" 10 times and "water" 10 times.
I'm aware that it all depends on the tf-idf value of "poland" and "water" but theoretically on an even playing field, would the algorithm help bring documents to the top of the results more if there were multiple terms in the document, or is it really term independent?
It is term independent. Remember, the tf-idf weighing scheme treats the query as a bag of words and each document is seen as a vector. For the above example, consider tf for poland is 100 while its idf is 1 in doc x. Also, consider tf for poland is 10 and tf for water is 2 is doc y. the idf of water is 1.
score of doc x = 100
score of doc y = 12
doc x ranked higher even though has one term.
its term independent. Depends on the ratio of how many documents contain poland and how many contain water. it that ratio. If its half-half, than the second document wins. If the ratio is 100:1, then the first document wins since the ratio is more similar to in-document distribution of the words.

Resources