Which natural language processing problems are NP-Complete or NP-Hard?

I've searched the and tags (and related complexity tags), but have not turned up any results.

None of the NLP questions that are recommended are helpful, the closest are the following:

The Wikipedia list of NP-complete problems does not list any complexity results for NLP.

The only lead I've found is the paper Theoretical and Effective Complexity in Natural Language Processing by J. Morin (1995).

Any help or pointers is appreciated!

  • 1
    $\begingroup$ This seems too broad a question to me. NLP offers a wide range of problems covering everything from trivial to undecidable. Which specific class of tasks are you interested in? What kind of "lead" does that paper contain? $\endgroup$
    – Raphael
    Nov 27 '14 at 14:05
  • $\begingroup$ this is indeed rather surprisingly broad given the history of CS parsing tied up with NLP (loosely defined) via Chomsky theory/ parsing! ie CFG parsing could arguably be tied up with early NLP theory. but a more careful/ limited/ narrow/ strict definition of NLP to be more modern etc might help it. also, there is a lot of statistical analysis, probability, and imprecision in NLP such that exact decision problems may not be a overly meaningful way to model its complexity. it falls more into machine learning where complexity theory is not always so applicable... $\endgroup$
    – vzn
    Nov 27 '14 at 15:32
  • $\begingroup$ as a different/ novel direction in the literature see also AI complete $\endgroup$
    – vzn
    Nov 27 '14 at 15:36

LFG (Lexical-Functional Grammar) recognition is NP-Complete.

Edit per request: Lexical-Functional Grammar (LFG) [1] is a theory of natural language syntax, developed as an alternative to Chomsky's theories of transformational syntax. Some versions of Chomsky's theories are computationally equivalent to Unrestricted Grammars. LFG by contrast provides a grammar formalism which consists of a context-free grammar augmented by a feature system.

It's the feature system that's NP-complete. The proof works basically by noticing first that the feature system is at least as powerful as propositional logic, and second that grammaticality rests on satisfying all the propositional constraints governing the sentence. So it's the Satisfiability Problem hiding under another guise.

[1] "Lexical-Functional Grammar: A Formal System for Grammatical Representation" by Ronald M Kaplan and Joan Bresnan. The paper originally appeared in The Mental Representation of Grammatical Relations, ed. Joan Bresnan (Cambridge, MA: The MIT Press, 1982).

  • 2
    $\begingroup$ Please elaborate so that the answer can stand on its own. What are LFGs? How does the proof work, roughly? Are there any published references? $\endgroup$
    – Raphael
    Nov 27 '14 at 14:06

Maybe one should first define what is a natural language processing (NLP) problem.

For example, Context-Free (CF) grammars and languages were introduced by linguists (Chomsky type 2 language, work of Bar-Hillel and others). Ambiguity is a major problem in Linguistics for real sentence analysis, and in the formal study of CF grammars (ambiguity) and languages (inherent ambiguity). Ambiguity of a grammar is only semi-decidable.

So I guess the problem of ambiguity should be an answer to your question. Does it classify as a NLP problem?

Now if you take some modern formalizations of syntax, such as a CF backbone with feature structures (i.e. structured attributes), you quickly get Turing power (cf LFG which have been proved NP hard, or even Turing complete, depending on variants). So , if you are not careful, you have all the complexity problems you might dream about.

For more, you can also look at this question from SE-Linguistics: "Does the P versus NP conjecture in computer science have any direct relevance to linguistics?"

In my own answer, I actually criticize the meaningfulness of the question, or at least of some of its interpretations. Many of the problems that are considered in linguistics, regarding analysis of sentences, for translation or other purposes, are small problem, to be solved in a very short time. Some linguists might even dispute that there is real recursion in language structure, since whatever recursion there is is rarely very deep. Hence one might wonder about the linguistic relevance of complexity analysis which is defined asymptotically. The first question should be whether we ever get close enough to the asymptote for asymptotic analysis to be meaningful.

However, this remark does not apply to some aspect of NLP, when massive amount of data have to be processed. I know of at least two cases:

  • data mining in large corpora.

  • the inverse problem of linguistics: analysis of large corpora to mechanically extract the data characterizing a language, both structurally and to produce extensive lists of constituents, such as phonemes, vocabulary for various parts of speech (aka preterminals), prefixes and suffixes, or inflection mechanisms, to give a few examples.

I am no expert in data mining, and thus do not know whether it actually raises complexity problems related to the size of the corpora being processed. In that case, asymptotic complexity would indeed be an issue. But if it is mostly composed of a large number of small additive tasks, then it is more doubtful that asymptotic complexity matters much. However, I would imagine that some data mining techniques will work with correlations between independent documents, and that should raise corpus dependent complexity issues.

In the case of the inverse problem of linguistics, the identification of a language (which, I guess, could be considered a data mining problem), we are indeed trying to extract information by correlating all parts of large corpora. Then asymptotic complexity becomes extremely relevant. I have unfortunately no specific problem in mind, probably because such systems have a pragmatic aim, and people developing them will tend to simply avoid any form of higher complexity, quadratic being probably already beyond the available resources. But a search of the literature would probably raise some complexity issues.

Another point is that linguistics does not have clearcut laws like physics. It is more a matter of being close enough to what might be considered current linguistic consensus, since no two people speak exactly the same language. Hence, good approximations are usually sufficient when the aim is so elusive. The techniques I have seen were mainly fix-point techniques to identify parameters by iterative recomputation of some function based on the corpus structure, until it no longer makes much difference (plus user input to weed out remaining pathological cases).

Analysing properties of grammars and other formalized linguistic structures may also be a source of high complexity problems, as mentioned above for ambiguity, since natural language descriptions are usually large enough for asymptotic analysis to be meaningful.


as in my comment, sometimes P/NP complexity while quite powerful can be a hammer that makes all complexity questions look like nails, and in the field of AI & language translation with statistical, probabilistic, imprecision, and machine learning aspects, it may not be the ideal measurement at times, and theoretical computational complexity is not always considered central or relevant esp in more applied AI/ machine learning. in some ways the entire field has a more empirical aspect to measuring applied problem complexity. however, here is one angle not yet pointed out in other answers, there are some P/NP considerations in NLP language translation. eg these two papers


Not the answer you're looking for? Browse other questions tagged or ask your own question.