I have been searching and attempting to implement a word embedding model to predict similarity between words. I have a dataset made up 3,550 company names, the idea is that the user can provide a new word (which would not be in the vocabulary) and calculate the similarity between the new name and existing ones.
During preprocessing I got rid of stop words and punctuation (hyphens, dots, commas, etc). In addition, I applied stemming and separated prefixes with the hope to get more precision. Then words such as BIOCHEMICAL ended up as BIO CHEMIC which is the word divided in two (prefix and stem word)
The average company name length is made up 3 words with the following frequency:
The tokens that are the result of preprocessing are sent to word2vec:
#window: Maximum distance between the current and predicted word within a sentence
#min_count: Ignores all words with total frequency lower than this.
#workers: Use these many worker threads to train the model
#sg: The training algorithm, either CBOW(0) or skip gram(1). Default is 0s
word2vec_model = Word2Vec(prepWords,size=300, window=2, min_count=1, workers=7, sg=1)
After the model included all the words in the vocab , the average sentence vector is calculated for each company name:
df['avg_vector']=df2.apply(lambda row : avg_sentence_vector(row, model=word2vec_model, num_features=300, index2word_set=set(word2vec_model.wv.index2word)).tolist())
Then, the vector is saved for further lookups:
##Saving name and vector values in file
df.to_csv('name-submission-vectors.csv',encoding='utf-8', index=False)
If a new company name is not included in the vocab after preprocessing (removing stop words and punctuation), then I proceed to create the model again and calculate the average sentence vector and save it again.
I have found this model is not working as expected. As an example, calculating the most similar words pet is getting the following results:
ms=word2vec_model.most_similar('pet')
('fastfood', 0.20879755914211273)
('hammer', 0.20450574159622192)
('allur', 0.20118337869644165)
('wright', 0.20001833140850067)
('daili', 0.1990675926208496)
('mgt', 0.1908089816570282)
('mcintosh', 0.18571510910987854)
('autopart', 0.1729743778705597)
('metamorphosi', 0.16965581476688385)
('doak', 0.16890916228294373)
In the dataset, I have words such as paws or petcare, but other words are creating relationships with pet word.
This is the distribution of the nearer words for pet:
On the other hand, when I used the GoogleNews-vectors-negative300.bin.gz, I could not add new words to the vocab, but the similarity between pet and words around was as expected:
ms=word2vec_model.most_similar('pet')
('pets', 0.771199643611908)
('Pet', 0.723974347114563)
('dog', 0.7164785265922546)
('puppy', 0.6972636580467224)
('cat', 0.6891531348228455)
('cats', 0.6719794869422913)
('pooch', 0.6579219102859497)
('Pets', 0.636363685131073)
('animal', 0.6338439583778381)
('dogs', 0.6224827170372009)
This is the distribution of the nearest words:
I would like to get your advice about the following:
Is this dataset appropriate to proceed with this model?
Is the length of the dataset enough to allow word2vec "learn" the relationships between the words?
What can I do to improve the model to make word2vec create relationships of the same type as GoogleNews where for instance word pet is correctly set among similar words?
Is it feasible to implement another alternative such as fasttext considering the nature of the current dataset?
Do you know any public dataset that can be used along with the current dataset to create those relationships?
Thanks
3500 texts (company names) of just ~3 words each is only around 10k total training words, with a much smaller vocabulary of unique words.
That's very, very small for word2vec & related algorithms, which rely on lots of data, and sufficiently-varied data, to train-up useful vector arrangements.
You may be able to squeeze some meaningful training from limited data by using far more training epochs than the default epochs=5, and far smaller vectors than the default size=100. With those sorts of adjustments, you may start to see more meaningful most_similar() results.
But, it's unclear that word2vec, and specifically word2vec in your averaging-of-a-name's-words comparisons, is matched to your end goals.
Word2vec needs lots of data, doesn't look at subword units, and can't say anything about word-tokens not seen during training. An average-of-many-word-vectors can often work as an easy baseline for comparing multiword texts, but might also dilute some word's influence compared to other methods.
Things to consider might include:
Word2vec-related algorithms like FastText that also learn vectors for subword units, and can thus bootstrap not-so-bad guess vectors for words not seen in training. (But, these are also data hungry, and to use on a small dataset you'd again want to reduce vector size, increase epochs, and additionally shrink the number of buckets used for subword learning.)
More sophisticated comparisons of multi-word texts, like "Word Mover's Distance". (That can be quite expensive on longer texts, but for names/titles of just a few words may be practical.)
Finding more data that's compatible with your aims for a stronger model. A larger database of company names might help. If you just want your analysis to understand English words/roots, more generic training texts might work too.
For many purposes, a mere lexicographic comparison - edit distances, count of shared character-n-grams – may be helpful too, though it won't detect all synonyms/semantically-similar words.
Word2vec does not generalize to unseen words.
It does not even work well for wards that are seen but rare. It really depends on having many many examples of word usage. Furthermore a you need enough context left and right, but you only use company names - these are too short. That is likely why your embeddings perform so poorly: too little data and too short texts.
Hence, it is the wrong approach for you. Retraining the model with the new company name is not enough - you still only have one data point. You may as well leave out unseen words, word2vec cannot work better than that even if you retrain.
If you only want to compute similarity between words, probably you don't need to insert new words in your vocabulary.
By eye, I think you can also use FastText without the need to stem the words. It also computes vectors for unknown words.
From FastText FAQ:
One of the key features of fastText word representation is its ability
to produce vectors for any words, even made-up ones. Indeed, fastText
word vectors are built from vectors of substrings of characters
contained in it. This allows to build vectors even for misspelled
words or concatenation of words.
FastText seems to be useful for your purpose.
For your task, you can follow FastText supervised tutorial.
If your corpus proves to be too small, you can build your model starting from availaible pretrained vectors (pretrainedVectors parameter).
I've been reading some NLP with Deep Learning papers and found Fine-tuning seems to be a simple but yet confusing concept. There's been the same question asked here but still not quite clear.
Fine-tuning pre-trained word embeddings to task-specific word embeddings as mentioned in papers like Y. Kim, “Convolutional Neural Networks for Sentence Classification,” and K. S. Tai, R. Socher, and C. D. Manning, “Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks,” had only a brief mention without getting into any details.
My question is:
Word Embeddings generated using word2vec or Glove as pretrained word vectors are used as input features (X) for downstream tasks like parsing or sentiment analysis, meaning those input vectors are plugged into a new neural network model for some specific task, while training this new model, somehow we can get updated task-specific word embeddings.
But as far as I know, during the training, what back-propagation does is updating the weights (W) of the model, it does not change the input features (X), so how exactly does the original word embeddings get fine-tuned? and where do these fine-tuned vectors come from?
Yes, if you feed the embedding vector as your input, you can't fine-tune the embeddings (at least easily). However, all the frameworks provide some sort of an EmbeddingLayer that takes as input an integer that is the class ordinal of the word/character/other input token, and performs a embedding lookup. Such an embedding layer is very similar to a fully connected layer that is fed a one-hot encoded class, but is way more efficient, as it only needs to fetch/change one row from the matrix on both front and back passes. More importantly, it allows the weights of the embedding to be learned.
So the classic way would be to feed the actual classes to the network instead of embeddings, and prepend the entire network with a embedding layer, that is initialized with word2vec / glove, and which continues learning the weights. It might also be reasonable to freeze them for several iterations at the beginning until the rest of the network starts doing something reasonable with them before you start fine tuning them.
One hot encoding is the base for constructing initial layer for embeddings. Once you train the network one hot encoding essentially serves as a table lookup. In fine-tuning step you can select data for specific works and mention variables that need to be fine tune when you define the optimizer using something like this
embedding_variables = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope="embedding_variables/kernel")
ft_optimizer = tf.train.AdamOptimizer(learning_rate=0.001,name='FineTune')
ft_op = ft_optimizer.minimize(mean_loss,var_list=embedding_variables)
where "embedding_variables/kernel" is the name of the next layer after one-hot encoding.
I am working on a research project on text data (it's about search engine queries supervised classification). I have already implemented different methods and I have also used different models for the text (such as binary vectors of the dimention of my vocabulary - 1 if the i-th word appears in the text, 0 otherwise - or words embedding with the model word2vec).
My advisor told me that maybe we could find another representation of the queries using Recurrent Neural Network. This representation should keep into account the sequentiality of the words in the text thanks to the recurrence relation. I have read some documentation about RNN but I haven't find anything useful for this goal. I have read lot of things about language modelling (which predict probabilities of the words), but I don't understand how I could adapt this model in order to obtain something like an embedded vector.
Thank you very much!
Usually, if one wants to obtain embeddings from a query or a sentence exploiting RNN, the logits are used. The logits are simply the output values of the network after the forward pass of the full sentence/query.
The logit values produce a vector that has the dimensions of the output layer (i.e. number of the target classes): usually, it is the vocabulary, since they are extracted from a language model.
For hints have a look at these:
http://arxiv.org/abs/1603.07012
How does word2vec give one hot word vector from the embedding vector?
Note that in principle one could use also use bidirectional networks or networks trained on other tasks, obtaining smaller embeddings, even if this last option is kind of fancy and it has not been explored up to my knowledge.
I'm working on a text classification problem, and I have problems with missing values on some features.
I'm calculating class probabilities of words from labeled training data.
For example;
Let word foo belongs to class A for 100 times and belongs to class B for 200 times. In this case, i find class probability vector as [0.33,0.67] , and give it along with the word itself to classifier.
Problem is that, in the test set, there are some words that have not been seen in training data, so they have no probability vectors.
What could i do for this problem?
I ve tried giving average class probability vector of all words for missing values, but it did not improve accuracy.
Is there a way to make classifier ignore some features during evaluation just for specific instances which does not have a value for giving feature?
Regards
There is many way to achieve that
Create and train classifiers for all sub-set of feature you have. You can train your classifier on sub-set with the same data as tre training of the main classifier.
For each sample juste look at the feature it have and use the classifier that fit him the better. Don't try to do some boosting with thoses classifiers.
Just create a special class for samples that can't be classified. Or you have experimented result too poor with so little feature.
Sometimes humans too can't succefully classify samples. In many case samples that can't be classified should just be ignore. The problem is not in the classifier but in the input or can be explain by the context.
As nlp point of view, many word have a meaning/usage that is very similare in many application. So you can use stemming/lemmatization to create class of words.
You can also use syntaxic corrections, synonyms, translations (does the word come from another part of the world ?).
If this problem as enouph importance for you then you will end with a combination of the 3 previous points.
I have a very fundamental question. I have two sets of documents, one for training and one for testing. I would like to train a Logistic regression classifier with the training documents. I want to know if I'm doing the right thing.
First find the list of all unique words in the training document and call it vocabulary.
For each word in the vocabulary, find its TFIDF in every training document. A document is then represented as vector of these TFIDF scores.
My question is:
1. How do I represent the test documents? Say, one of the test documents does not have any word that is in the vocabulary. In that case , the TFIDF scores will be zero for all words in the vocabulary for that document.
I'm trying to use LIBSVM which uses the sparse vector format. For the case of the above document, which has all entries set to 0 in its vector representation, how do I represent it?
You have to store enough information about the training corpus to do the TF IDF transform on unseen documents. This means you'll need the document frequencies of the terms in the training corpus. Ignoring unseen words in test docs is fine. Your svm won't learn a weight for them anyway. Note that unseen terms should be rare in the test corpus if your training and test distributions are similar. So even if a few terms are dropped, you'll still have plenty of terms to classify the doc.