At least 3 types of n-grams can be considered for representing text documents:
byte-level n-grams
character-level n-grams
word-level n-grams
It's unclear to me which one should be used for a given task (clustering, classification, etc). I read somewhere that character-level n-grams are preferred to word-level n-grams when the text contains typos, so that "Mary loves dogs" remains similar to "Mary lpves dogs".
Are there other criteria to consider for choosing the "right" representation?
Evaluate. The criterion for choosing the representation is whatever works.
Indeed, character level (!= bytes, unless you only care about english) probably is the most common representation, because it is robust to spelling differences (which do not need to be errors, if you look at history; spelling changes). So for spelling correction purposes, this works well.
On the other hand, Google Books n-gram viewer uses word level n-grams on their books corpus. Because they don't want to analyze spelling, but term usage over time; e.g. "child care", where the individual words aren't as interesting as their combination. This was shown to be very useful in machine translation, often referred to as "refrigerator magnet model".
If you are not processing international language, bytes may be meaningful, too.
I would outright discard byte-level n-grams for text-related tasks, because bytes are not a meaningful representation of anything.
Of the 2 remaining levels, the character-level n-grams will need much less storage space and will , subsequently, hold much less information. They are usually utilized in such tasks as language identification, writer identification (i.e. fingerprinting), anomaly detection.
As for word-level n-grams, they may serve the same purposes, and much more, but they need much more storage. For instance, you'll need up to several gigabytes to represent in memory a useful subset of English word 3-grams (for general-purpose tasks). Yet, if you have a limited set of texts you need to work with, word-level n-grams may not require so much storage.
As for the issue of errors, a sufficiently large word n-grams corpus will also include and represent them. Besides, there are various smoothing methods to deal with sparsity.
There other issue with n-grams is that they will almost never be able to capture the whole needed context, so will only approximate it.
You can read more about n-grams in the classic Foundations of Statistical Natural Language Processing.
I use character ngrams on small strings, and word ngrams for something like text classification of larger chunks of text. It is a matter of which method will preserve the context you need more or less...
In general for classification of text, word ngrams will help a bit with word-sense dissambiguation, where character ngrams would be easily confused and your features could be completely ambiguous. For unsupervised clustering, it will depend on how general you want your clusters, and on what basis you want docs to converge. I find stemming, stopword removal, and word bigrams work well in unsupervised clustering tasks on fairly large corpora.
Character ngrams are great for fuzzy string matching of small strings.
I like to think of a set of grams as a vector, and imagine comparing vectors with the grams you have, then ask yourself if what you are comparing maintains enough context to answer the question you are trying to answer.
HTH
Related
I have just started a project in NLP. Suppose I have a graph for each word that shows the polarity distribution of sentiments for that word in different sentences. I want to know what I can use to recognize the feelings of new words? Any other use you have in mind I will be happy to share.
I apologize for any possible errors in my writing. Thanks a lot
Assuming you've got some words that have been hand-labeled with positive/negative sentiments, but then you encounter some new words that aren't labeled:
If you encounter the new words totally alone, outside of contexts, there's not much you can do. (Maybe, you could go out to try to find extra texts with those new words, such as vis dictionaries or the web, then use those larger texts in the next approach.)
If you encounter the new words inside texts that also include some of your hand-labeled words, you could try guessing that the new words are most like the words you already know that are closest-to, or used-in-the-same-places. This would leverage what's called "the distributional hypothesis" – words with similar distributions have similar meanings – that underlies a lot of computer natural-language analysis, including word2vec.
One simple thing to try along these lines: across all your texts, for every unknown word U, tally up the counts all neighboring words within N positions. (N could be 1, or larger.) From that, pick the top 5 words occuring most often near the unknown word, and look up your prior labels, and avergae them together (perhaps weighted by the number of occurrences.)
You'll then have a number for the new word.
Alternatively, you could train a word2vec set-of-word-vectors for all of your texts, including the unknown & know words. Then, ask that model for the N most-similar neighbors to your unknown word. (Again, N could be small or large.) Then, from among those neighbors with known labels, average them together (again perhaps weighted by similarity), to get a number for the previously unknown word.
I wouldn't particularly expect either of these techniques to work very well. The idea that individual words can have specific sentiment is somewhat weak given the way that in actual language, their meaning is heavily modified, or even reversed, by the surrounding grammar/context. But in each case these simple calculate-from-neighbors techniqyes are probably better than random guesses.
If your real aim is to calculate the overall sentiment of longer texts, like sentences, paragraphs, reviews, etc, then you should discard your labels of individual words an acquire/create labels for full texts, and apply real text-classification techniques to those larger texts. A simple word-by-word approach won't do very well compared to other techniques – as long as those techniques have plenty of labeled training data.
I have look into some word embedding techniques, such as
CBOW: from context to single word. Weight matrix produced used as embedding vector
Skip gram: from word to context (from what I see, its acutally word to word, assingle prediction is enough). Again Weight matrix produced used as embedding
Introduction to these tools would always quote "cosine similarity", which says words of similar meanning would convert to similar vector.
But these methods all based on the 'context', account only for words around a target word. I should say they are 'syntagmatic' rather than 'paradigmatic'. So why the close in distance in a sentence indicate close in meaning? I can think of many counter example that frequently occurs
"Have a good day". (good and day are vastly different, though close in distance).
"toilet" "washroom" (two words of similar meaning, but a sentence contains one would unlikely to contain another)
Any possible explanation?
This sort of "why" isn't a great fit for StackOverflow, but some thoughts:
The essence of word2vec & similar embedding models may be compression: the model is forced to predict neighbors using far less internal state than would be required to remember the entire training set. So it has to force similar words together, in similar areas of the parameter space, and force groups of words into various useful relative-relationships.
So, in your second example of 'toilet' and 'washroom', even though they rarely appear together, they do tend to appear around the same neighboring words. (They're synonyms in many usages.) The model tries to predict them both, to similar levels, when typical words surround them. And vice-versa: when they appear, the model should generally predict the same sorts of words nearby.
To achieve that, their vectors must be nudged quite close by the iterative training. The only way to get 'toilet' and 'washroom' to predict the same neighbors, through the shallow feed-forward network, is to corral their word-vectors to nearby places. (And further, to the extent they have slightly different shades of meaning – with 'toilet' more the device & 'washroom' more the room – they'll still skew slightly apart from each other towards neighbors that are more 'objects' vs 'places'.)
Similarly, words that are formally antonyms, but easily stand-in for each-other in similar contexts, like 'hot' and 'cold', will be somewhat close to each other at the end of training. (And, their various nearer-synonyms will be clustered around them, as they tend to be used to describe similar nearby paradigmatically-warmer or -colder words.)
On the other hand, your example "have a good day" probably doesn't have a giant influence on either 'good' or 'day'. Both words' more unique (and thus predictively-useful) senses are more associated with other words. The word 'good' alone can appear everywhere, so has weak relationships everywhere, but still a strong relationship to other synonyms/antonyms on an evaluative ("good or bad", "likable or unlikable", "preferred or disliked", etc) scale.
All those random/non-predictive instances tend to cancel-out as noise; the relationships that have some ability to predict nearby words, even slightly, eventually find some relative/nearby arrangement in the high-dimensional space, so as to help the model for some training examples.
Note that a word2vec model isn't necessarily an effective way to predict nearby words. It might never be good at that task. But the attempt to become good at neighboring-word prediction, with fewer free parameters than would allow a perfect-lookup against training data, forces the model to reflect underlying semantic or syntactic patterns in the data.
(Note also that some research shows that a larger window influences word-vectors to reflect more topical/domain similarity – "these words are used about the same things, in the broad discourse about X" – while a tiny window makes the word-vectors reflect a more syntactic/typical similarity - "these words are drop-in replacements for each other, fitting the same role in a sentence". See for example Levy/Goldberg "Dependency-Based Word Embeddings", around its Table 1.)
‘Embedding’ mean a semantic vector representation. e.g. how to represent words such that synonyms are nearer than antonyms or other unrelated words.
Embeddings algorithms like Word2vec maps entities be it e-commerce
items or words (say in English language), to N-dimensional vectors.
Now since you have a mathematical representation of the entities in
a Euclidean space, you can use associated semantics such as distance
between vectors. e.g:
For a given item say ‘Levis Jeans’ recommend the most related items
which are often co-purchased with it.
This can be easily done: search the nearest vectors to the vector of
‘Levis Jeans’, and recommend them. You will find that the nearest
vectors correspond to items such as T-shirts etc., which are
relevant to the Levis Jeans. Similarly it preserves
distance/similarity between words e.g.: King - Queen = Man - Woman !
Yes, Word2vec captures such co-occurrance relationships, when
mapping the items/words to vectors also called as ‘item/word
embeddings’.
This is not specifically targeted to sentence embeddings but nevertheless here you get some crucial insights extremely relevant to the core logic behind embedding generation. Read till the end.
I have some questions about Word2Vec:
What determines the dimension of the result model vectors?
What is elements of this vectors?
Can I use Word2Vec for polysemy solving problems (state = administrative unit vs state = condition), if I already have texts for every meaning of words?
(1) You pick the desired dimensionality, as a meta-parameter of the model. Rigorous projects with enough time may try different sizes, to see what works best for their qualitative evaluations.
(2) Individual dimensions/elements of each word-vector (floating-point numbers), in vanilla word2vec are not easily interpretable. It's only the arrangement of words as a whole that has usefulness – placing similar words near each other, and making relative directions (eg "towards 'queen' from 'king'") match human intuitions about categories/continuous-properties. And, because the algorithms use explicit randomization, and optimized multi-threaded operation introduces thread-scheduling randomness to the order-of-training-examples, even the exact same data can result in different (but equally good) vector-coordinates from run-to-run.
(3) Basic word2vec doesn't have an easy fix, but there's a bunch of hints of polysemy in the vectors, and research work to do more to disambiguate contrasting senses.
For example, generally more-polysemous word-tokens wind up with word-vectors that are some combination of their multiple senses, and (often) of a smaller-magnitude than less-polysemous words.
This early paper used multiple representations per word to help discover polysemy. Similar later papers like this one use clustering-of-contexts to discover polysemous words then relabel them to give each sense its own vector.
This paper manages an impressive job of detecting alternate senses via postprocessing of normal word2vec vectors.
I would like to be able to curate definitional summaries for a specific term from a textbook.
For example, from a Biology textbook, I would like to be able form a concise summary for the word "mitochondria". I have tried this by first parsing through the textbook for all sentences that contain the word "mitochondria", and feeding those sentences through summarization algorithms such as TextRank and LexRank, but those algorithms were not able to determine "definitional" sentences that well.
By definitional summaries, I mean useful sentences as far as a definition goes. For example, the sentence "The mitochondria is the powerhouse of the cell" would be a definitional sentence while the sentence "Fungal cells also contain mitochondria and a complex system of internal membranes, including the endoplasmic reticulum and Golgi apparatus" is not really pertinent to the definition of the mitochondria.
Any help or leads would be very much appreciated
There isn't really a straightforward way to do this, but you do have some options:
Just use a regex for "mitochondria is". It is the stupidest possible thing, but given a textbook it might prove satisfactory. It's simple enough testing should be easy, and at worst provides a baseline to compare alternatives to.
Run a parser (example: Stanford Parser) on each sentence with the word "mitochondria", and extract sentences where mitochondria is the subject. This would eliminate the negative example you gave. You would have to tune this, perhaps restricting main verbs, accounting for coordinators, and so on.
Use Information Extraction (example: Stanford OpenIE) to get a list of facts about mitochondria (like is-in(mitochondria, cell)) and do something with that.
This is a very open ended question. I can try to point how I would approach this...
One way would be to use some kind of vector representation for text (word2vec
or sent2vec come to mind).
Then by encoding the average of the sentences in vector format and checking the cosine similarity of this and of the term you seek, you could be getting something close to the definitional sentences you seek.
Even testing the cosine similarity of the averaged sentences you get out of the summary algorithm and the term might get you close to judge how close you are
In the part of speech tagger, the best probable tags for the given sentence is determined using HMM by
P(T*) = argmax P(Word/Tag)*P(Tag/TagPrev)
T
But when 'Word' did not appear in the training corpus, P(Word/Tag) produces ZERO for given all possible tags, this leaves no room for choosing the best.
I have tried few ways,
1) Assigning small amount of probability for all unknown words, P(UnknownWord/AnyTag)~Epsilon... means this completely ignores the P(Word/Tag) for unknowns word by assigning the constant probability.. So decision making on unknown word is by prior probabilities.. As expected it is not producing good result.
2) Laplace Smoothing
I confused with this. I don't know what is difference between (1) and this. My way of understanding Laplace Smoothing adds the constant probability(lambda) to all unknown & Known words.. So the All Unknown words will get constant probability(fraction of lambda) and Known words probabilities will be the same relatively since all word's prob increased by Lambda.
Is the Laplace Smoothing same as the previous one ?
*)Is there any better way of dealing with unknown words ?
Your two approaches are similar, but, if I understand correctly, they differ in one key way. In (1) you are assigning extra mass to counts of unknown words and in (2) you are assigning extra mass to all counts. You definitely want to do (2) and not (1).
One of the problems with Laplace smoothing is that it give too much of a boost to unknown words and drags down the probabilities of high-probability words too much (relatively speaking). Your version (1) would actually worsen this problem. Basically, it would over-smooth.
Laplace smoothing words ok for an HMM, but it's not great. Most people do add-one smoothing but you could experiment with things like add-one-half or whatever.
If you want to move beyond this naive approach to smoothing, check out "one-count smoothing", as described in the Appendix of Jason Eisner's HMM tutorial. The basic idea here is that for unknown words more probability mass should be given to tags that appear with a wider variety of low frequency words. For example, since the tag NOUN appears on a large number of different words and DETERMINER appears on a small number of different words, it is more likely that an unseen word will be a NOUN.
If you want to get even fancier, you could use a Chinese Restaurant Process model taken from non-parametric Bayesian statistics to put a prior distribution on unseen word/tag combinations. Kevin Knight's Bayesian inference tutorial has details.
I think the HMM-based TnT tagger provides a better approach to handle unknown words (see the approach in TnT tagger's paper).
The accuracy results (for known words and unknown words) of TnT and other two POS and morphological taggers on 13 languages including Bulgarian, Czech, Dutch, English, French, German, Hindi, Italian, Portuguese, Spanish, Swedish, Thai and Vietnamese, can be found in this article.