I have a DB containing tf-idf vectors of about 30,000 documents.
I would like to return for a given document a set of similar documents - about 4 or so.
I thought about implementing a K-Means (clustering algorithm) on the data (with cosine similarity), but I don't know whether it's the best choice because of many uncertainties: I'm not sure what to put in my initial clusters, I don't know how many clusters to create, I fear the clusters will be too unbalanced, I'm not sure the results quality will be good, etc.
Any advice and help from experienced users will be greatly appreciated.
Thank you,
Katie
I would like to return for a given document a set of similar documents - about 4 or so.
Then don't do k-means. Just return the four closest documents by tf-idf similarity, as any search engine would do. You can implement this as a k-nearest neighbor search, or more easily by installing a search engine library and using the initial document as a query. Lucene comes to mind.
If I understand, you
read 30k records from a bigger db to a cache file / to memory
cosine similarity, 10 terms * 30k records -> best 4.
Can you estimate the runtimes of these phases separately ?
read or cache: how often will this be done,
how big are the 30k vectors all together ?
10 * 30k multiply-adds: in your c / java / ... or in some opaque db ?
In c or java, that should take < 1 second.
In general, make some back-of-the-envelope estimates
before getting fancy.
(By the way,
I find best-4 faster and simpler in straight-up c than std::partial_sort; ymmv.)
Related
I have a collection with 100 million documents of geometry.
I have a second collection with time data associated to each of the other geometries. This will be 365 * 96 * 100 million or 3.5 trillion documents.
Rather than store the 100 million entries (365*96) times more than needed, I want to keep them in separate collections and do a type of JOIN/DBRef/Whatever I can in MongoDB.
First and foremost, I want to get a list of GUIDs from the geometry collection by using a geoIntersection. This will filter it down to 100 million to 5000. Then using those 5000 geometries guids I want to filter the 3.5 trillion documents based on the 5000 goemetries and additional date criteria I specify and aggregate the data and find the average. You are left with 5000 geometries and 5000 averages for the date criteria you specified.
This is basically a JOIN as I know it in SQL, is this possible in MongoDB and can it be done optimally in say less than 10 seconds.
Clarify: as I understand, this is what DBrefs is used for, but I read that it is not efficient at all, and with dealing with this much data that it wouldn't be a good fit.
If you're going to be dealing with a geometry and its time series data together, it makes sense to store them in the same doc. A years worth of data in 15 minute increments isn't killer - and you definitely don't want a document for every time-series entry! Since you can retrieve everything you want to operate on as a single geometry document, it's a big win. Note that this also let's you sparse things up for missing data. You can encode the data differently if it's sparse rather than indexing into a 35040 slot array.
A $geoIntersects on a big pile of geometry data will be a performance issue though. Make sure you have some indexing on (like 2dsphere) to speed things up.
If there is any way you can build additional qualifiers into the query that could cheaply eliminate members from the more expensive search, you may make things zippier. Like, say the search will hit states in the US. You could first intersect the search with state boundaries to find the states containing the geodata and use something like a postal code to qualify the documents. That would be a really quick pre-search against 50 documents. If a search boundary was first determined to hit 2 states, and the geo-data records included a state field, you just winnowed away 96 million records (all things being equal) before the more expensive geo part of the query. If you intersect against smallish grid coordinates, you may be able to winnow it further before the geo data is considered.
Of course, going too far adds overhead. If you can correctly tune the system to the density of the 100 million geometries, you may be able to get the times down pretty low. But without actually working with the specifics of the problem, it's hard to know. That much data probably requires some specific experimentation rather than relying on a general solution.
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've heard quite a couple times people talking about KDB deal with millions of rows in nearly no time. why is it that fast? is that solely because the data is all organized in memory?
another thing is that is there alternatives for this? any big database vendors provide in memory databases ?
A quick Google search came up with the answer:
Many operations are more efficient with a column-oriented approach. In particular, operations that need to access a sequence of values from a particular column are much faster. If all the values in a column have the same size (which is true, by design, in kdb), things get even better. This type of access pattern is typical of the applications for which q and kdb are used.
To make this concrete, let's examine a column of 64-bit, floating point numbers:
q).Q.w[] `used
108464j
q)t: ([] f: 1000000 ? 1.0)
q).Q.w[] `used
8497328j
q)
As you can see, the memory needed to hold one million 8-byte values is only a little over 8MB. That's because the data are being stored sequentially in an array. To clarify, let's create another table:
q)u: update g: 1000000 ? 5.0 from t
q).Q.w[] `used
16885952j
q)
Both t and u are sharing the column f. If q organized its data in rows, the memory usage would have gone up another 8MB. Another way to confirm this is to take a look at k.h.
Now let's see what happens when we write the table to disk:
q)`:t/ set t
`:t/
q)\ls -l t
"total 15632"
"-rw-r--r-- 1 kdbfaq staff 8000016 May 29 19:57 f"
q)
16 bytes of overhead. Clearly, all of the numbers are being stored sequentially on disk. Efficiency is about avoiding unnecessary work, and here we see that q does exactly what needs to be done when reading and writing a column - no more, no less.
OK, so this approach is space efficient. How does this data layout translate into speed?
If we ask q to sum all 1 million numbers, having the entire list packed tightly together in memory is a tremendous advantage over a row-oriented organization, because we'll encounter fewer misses at every stage of the memory hierarchy. Avoiding cache misses and page faults is essential to getting performance out of your machine.
Moreover, doing math on a long list of numbers that are all together in memory is a problem that modern CPU instruction sets have special features to handle, including instructions to prefetch array elements that will be needed in the near future. Although those features were originally created to improve PC multimedia performance, they turned out to be great for statistics as well. In addition, the same synergy of locality and CPU features enables column-oriented systems to perform linear searches (e.g., in where clauses on unindexed columns) faster than indexed searches (with their attendant branch prediction failures) up to astonishing row counts.
Sources(S): http://www.kdbfaq.com/kdb-faq/tag/why-kdb-fast
as for speed, the memory thing does play a big part but there are several other things, fast read from disk for hdb, splaying etc. From personal experienoce I can say, you can get pretty good speeds from c++ provided you want to write that much code. With kdb you get all that and some more.
another thing about speed is also speed of coding. Steep learning curve but once you get it, complex problems can be coded in minutes.
alternatives you can look at onetick or google in memory databases
kdb is fast but really expensive. Plus, it's a pain to learn Q. There are a few alternatives such as DolphinDB, Quasardb, etc.
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?
I have to check performance of various clustering algos using different performance operators in rapidminer. For that I want to know the following things:
what does cluster number index value shows which is output of cluster count performance operator?
what does small and large value of avg within cluster distance and avg. within centroid distance mean in terms of good and bad clustering?
I also want to check other indexes value like Dunn index,Jaccard index, Fowlkes–Mallows for various clustering algos. but rapidminer don't have any operator for this, what to do for that. I don't have experience with R.
I have copied part of the answer I gave on the Rapid-I forum
The cluster number index is the count of clusters - pointless you might say but when used with DBSCAN, it can be quite interesting http://rapidminernotes.blogspot.co.uk/2010/12/counting-clusters.html
The avg within cluster and centroid distances are hard to interpret - one thing to search for is "elbow criterion" in this context. As the number of clusters varies, note how the validity measure changes and look for an "elbow" that marks the point where the natural progression of the measure dominates the structure.
R has many validity measures and it's worth investing some time because you can always call the R process from RapidMiner which makes it easier to work out what is going on.