I'm fairly new to clustering and related topics so please forgive my questions.
I'm trying to get introduced into this area by doing some tests, and as a first experiment I'd like to create clusters on tweets based on content similarity. The basic idea for the experiment would be storing tweets on a database and periodically calculate the clustering (ie. using a cron job). Please note that the database would obtain new tweets from time to time.
Being ignorant in this field, my idea (probably naive) would be to do something like this:
1. For each new tweet in the db, extract N-grams (N=3 for example) into a set
2. Perform Jaccard similarity and compare with each of the existing clusters. If result > threshold then it would be assigned to that cluster
3. Once finished I'd get M clusters containing similar tweets
Now I see some problems with this basic approach. Let's put aside computational cost, how would the comparison between a tweet and a cluster be done? Assuming I have a tweet Tn and a cluster C1 containing T1, T4, T10 which one should I compare it to? Given that we're talking about similarity, it could well happen that sim(Tn,T1) > threshold but sim(Tn,T4) < threshold. My gut feeling tells me that something like an average should be used for the cluster, in order to avoid this problem.
Also, it could happen that sim(Tn, C1) and sim(Tn, C2) are both > threshold but similarity with C1 would be higher. In that case Tn should go to C1. This could be done brute force as well to assign the tweet to the cluster with maximum similarity.
And last of all, it's the computational issue. I've been reading a bit about minhash and it seems to be the answer to this problem, although I need to do some more research on it.
Anyway, my main question would be: could someone with experience in the area recommend me which approach should I aim to? I read some mentions about LSA and other methods, but trying to cope with everything is getting a bit overwhelming, so I'd appreciate some guiding.
From what I'm reading a tool for this would be hierarchical clustering, as it would allow regrouping of clusters whenever new data enters. Is this correct?
Please note that I'm not looking for any complicated case. My use case idea would be being able to cluster similar tweets into groups without any previous information. For example, tweets from Foursquare ("I'm checking in ..." which are similar to each other would be one case, or "My klout score is ..."). Also note that I'd like this to be language independent, so I'm not interested in having to deal with specific language issues.
It looks like to me that you are trying to address two different problems in one, i.e. "syntactic" and "semantic" clustering. They are quite different problems, expecially if you are in the realm of short-text analysis (and Twitter is the king of short-text analysis, of course).
"Syntactic" clustering means aggregating tweets that come, most likely, from the same source. Your example of Foursquare fits perfectly, but it is also common for retweets, people sharing online newspaper articles or blog posts, and many other cases. For this type of problem, using a N-gram model is almost mandatory, as you said (my experience suggests that N=2 is good for tweets, since you can find significant tweets that have as low as 3-4 features). Normalization is also an important factor here, removing RT tag, mentions, hashtags might help.
"Semantic" clustering means aggregating tweets that share the same topic. This is a much more difficult problem, and it won't likely work if you try to aggregate random sample of tweets, due to the fact that they, usually, carry too little information. These techniques might work, though, if you restrict your domain to a specific subset of tweets (i.e. the one matching a keyword, or an hashtag). LSA could be useful here, while it is useless for syntactic clusters.
Based on your observation, I think what you want is syntactic clustering. Your biggest issue, though, is the fact that you need online clustering, and not static clustering. The classical clustering algorithms that would work well in the static case (like hierarchical clustering, or union find) aren't really suited for online clustering , unless you redo the clustering from scratch every time a new tweet gets added to your database. "Averaging" the clusters to add new elements isn't a great solution according to my experience, because you need to retain all the information of every cluster member to update the "average" every time new data gets in. Also, algorithms like hierarchical clustering and union find work well because they can join pre-existant clusters if a link of similarity is found between them, and they don't simply assign a new element to the "closest" cluster, which is what you suggested to do in your post.
Algorithms like MinHash (or SimHash) are indeed more suited to online clustering, because they support the idea of "querying" for similar documents. MinHash is essentially a way to obtain pairs of documents that exceed a certain threshold of similarity (in particular, MinHash can be considered an estimator of Jaccard similarity) without having to rely on a quadratic algorithm like pairwise comparison (it is, in fact, O(nlog(n)) in time). It is, though, quadratic in space, therefore a memory-only implementation of MinHash is useful for small collections only (say 10000 tweets). In your case, though, it can be useful to save "sketches" (i.e., the set of hashes you obtain by min-hashing a tweet) of your tweets in a database to form an "index", and query the new ones against that index. You can then form a similarity graph, by adding edges between vertices (tweets) that matched the similarity query. The connected components of your graph will be your clusters.
This sounds a lot like canopy pre-clustering to me.
Essentially, each cluster is represented by the first object that started the cluster.
Objects within the outer radius join the cluster. Objects that are not within the inner radius of at least one cluster start a new cluster. This way, you get an overlapping (non-disjoint!) quantization of your dataset. Since this can drastically reduce the data size, it can be used to speed up various algorithms.
However don't expect useful results from clustering tweets. Tweet data is just to much noise. Most tweets have just a few words, too little to define a good similarity. On the other hand, you have the various retweets that are near duplicates - but trivial to detect.
So what would be a good cluster of tweets? Can this n-gram similarity actually capture this?
Related
I'm looking at finding people with similar purchasing behaviors with a given person or group as a starting point for a market research problem.
I'm going to use vectors and represent every person and their habits as a vector and then compare these vectors to return the base person or group. I'd probably use Faiss. I believe KNN can be used too.
But I'm looking to see is if I can use other methods such as clustering methods like k-means clustering for such a question, and with the presence of a given person or group as the base. I thought the only way clustering algs would work is to first cluster the data, then return the group that the 'base person or group' falls into. However, this would be costly and probably not very accurate. But potentially this technique can be used to reduce the search space.
So, do you know of any other ways? (non-Machine Learning or Information Retrieval methods would be welcomed too :) )
I want to use clustering techniques for binary data analysis. I have collected the data through survey in which i asked the users to select exactly 20 features out of list of 94 product features. The columns in my data represents the 94 product features and the rows represents the participants. I am trying to cluster the similar users in different user groups based on the product features they selected. Each user cluster should also tell me the product features associated with each cluster. I am using some open source clustering tools like NCSS and JMP. I was trying to use fuzzy clustering technique for achieveing my goal but unfortunately these tools do not deal with binary data. Can you please suggest me which technique would really be appropriate for my tasks , also which online tool i can use for using the cluster analysis on my data? As beacuse of the time limitation, I am not looking to code myself and i am only looking for some open source tools that have all the functionality available in them which i can use as it is.
Clustering for binary data is not really well defined.
Rather than looking for some tool/function that may or may not work by trial and error, you should first try to answer a 'simple" question:
What is a good cluster, mathematically?
Vague terms not allowed. The next questions to answer then are: I) when is clustering A better than clustering B (I.e. how does the computer compute quality), and ii) how can this be found efficiently.
You won't get far if you don't understand what you are doing just by calling random functions...
Also, is clustering actually what you are looking for? Most of the time with binary data e.g. frequent itemset mining is the better choice.
I have a big doubt about the way I should cluster sets using MinHash together with the banding technique.
I assume everyone reading has a good knowledge of MinHash so I won't define most of the terms I'm using.
My goal is to use MinHash to cluster users according to the similarity of their signatures. In a local, non-banded settings this would be trivial: if their signature hash is the same, they go in the same cluster.
If we split signatures in bands and process them indipendently, I can treat a band as I said before and generate a group of clusters for every band. My question is: how should I aggregate these clusters? Just merge them if they have at least an element in common? Or should I do something different?
Thanks
MinHash is not really meant as standalone clustering algorithm. It is meant as a candidate filter for near-duplicate detection.
When looking for similar documents, you compute the minhashes to retrieve candidates. You then still need to check these candidates - they could be false positives!
The more signatures agree, the more likely they really match.
So if you consider the near-duplicate scenario again: if a is a near duplicate of b and b is a near duplicate of c, then a should also be a near duplicate of c. If this holds, you can throw all these matches (after verification) together. If it doesn't consider a hierarchical clustering like strategy to merge (or not merge) candidates.
I know there are ways to find synonyms either by using NLTK/pywordnet or Pattern package in python but it isn't solving my problem.
If there are words like
bad,worst,poor
bag,baggage
lost,lose,misplace
I am not able to capture them. Can anyone suggest me a possible way?
There have been numerous research in this area in past 20 years. Yes computers don't understand language but we can train them to find similarity or difference in two words with the help of some manual effort.
Approaches may be:
Based on manually curated datasets that contain how words in a language are related to each other.
Based on statistical or probabilistic measures of words appearing in a corpus.
Method 1:
Try Wordnet. It is a human-curated network of words which preserves the relationship between words according to human understanding. In short, it is a graph with nodes as something called 'synsets' and edges as relations between them. So any two words which are very close to each other are close in meaning. Words that fall within the same synset might mean exactly the same. Bag and Baggage are close - which you can find either by iteratively exploring node-to-node in a breadth first style - like starting with 'baggage', exploring its neighbors in an attempt to find 'baggage'. You'll have to limit this search upto a small number of iterations for any practical application. Another style is starting a random walk from a node and trying to reach the other node within a number of tries and distance. It you reach baggage from bag say, 500 times out of 1000 within 10 moves, you can be pretty sure that they are very similar to each other. Random walk is more helpful in much larger and complex graphs.
There are many other similar resources online.
Method 2:
Word2Vec. Hard to explain it here but it works by creating a vector of a user's suggested number of dimensions based on its context in the text. There has been an idea for two decades that words in similar context mean the same. e.g. I'm gonna check out my bags and I'm gonna check out my baggage both might appear in text. You can read the paper for explanation (link in the end).
So you can train a Word2Vec model over a large amount of corpus. In the end, you will be able to get 'vector' for each word. You do not need to understand the significance of this vector. You can this vector representation to find similarity or difference between words, or generate synonyms of any word. The idea is that words which are similar to each other have vectors close to each other.
Word2vec came up two years ago and immediately became the 'thing-to-use' in most of NLP applications. The quality of this approach depends on amount and quality of your data. Generally Wikipedia dump is considered good training data for training as it contains articles about almost everything that makes sense. You can easily find ready-to-use models trained on Wikipedia online.
A tiny example from Radim's website:
>>> model.most_similar(positive=['woman', 'king'], negative=['man'], topn=1)
[('queen', 0.50882536)]
>>> model.doesnt_match("breakfast cereal dinner lunch".split())
'cereal'
>>> model.similarity('woman', 'man')
0.73723527
First example tells you the closest word (topn=1) to words woman and king but meanwhile also most away from the word man. The answer is queen.. Second example is odd one out. Third one tells you how similar the two words are, in your corpus.
Easy to use tool for Word2vec :
https://radimrehurek.com/gensim/models/word2vec.html
http://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf (Warning : Lots of Maths Ahead)
I'm doing a little research on how to cluster articles into 'news stories' ala Google News.
Looking at previous questions here on the subject, I often see it recommended to simply pull out a vector of words from an article, weight some of the words more if they're in certain parts of the article (e.g. the headline), and then to use something like a k-means algorithm to cluster the articles.
But this leads to a couple of questions:
With k-means, how do you know in advance how much k should be? In a dynamic news environment you may have a very variable number of stories, and you won't know in advance how many stories a collection of articles represents.
With hierarchal clustering algorithms, how do you decide which clusters to use as your stories? You'll have clusters at the bottom of the tree that are just single articles, which you obviously won't want to use, and a cluster at the root of the tree which has all of the articles, which again you won't want...but how do you know which clusters in between should be used to represent stories?
Finally, with either k-means or hierarchal algorithms, most literature I have read seems to assume you have a preset collection of documents you want to cluster, and it clusters them all at once. But what of a situation where you have new articles coming in every so often. What happens? Do you have to cluster all the articles from scratch, now that there's an additional one? This is why I'm wondering if there are approaches that let you 'add' articles as you go without re-clustering from scratch. I can't imagine that's very efficient.
I worked on a start-up that built exactly this: an incremental clustering engine for news articles. We based our algorithm on this paper: Web Document Clustering Using Document Index Graph (http://ieeexplore.ieee.org/xpl/articleDetails.jsp?reload=true&arnumber=4289851). Worked well for us for 10K articles / day.
It has two main advantages:
1) It's incremental, which addresses the problem you have with having to deal with a stream of incoming articles (rather than clustering all at once)
2) It uses phrase-based modeling, as opposed to just "bag of words", which results in much higher accuracy.
A Google search pops up http://www.similetrix.com, they might have what you're looking for.
I would do a search for adaptive K-means clustering algorithms. There is a good section of research devoted to the problems you describe. Here is one such paper (pdf)