StackOverflow2013

Note that there are some explanatory texts on larger screens.

plurals

PO
primarykey
Id
9845516
data
AcceptedAnswerId
0
AnswerCount
0
ClosedDate
CommentCount
1
CommunityOwnedDate
CreationDate
2012-03-23T19:39:43.920
FavoriteCount
0
LastActivityDate
2012-03-23T19:45:37.490
LastEditDate
2012-03-23T19:45:37.490
LastEditorUserId
729627
OwnerUserId
729627
ParentId
9843424
PostTypeId
2
Score
2
ViewCount
0
LastEditorDisplayName
text
Body
Consider using a non-deterministic <a href="http://en.wikipedia.org/wiki/Trie" rel="nofollow">trie automaton</a> covering the full set of recognized suffixes, but analyzing the word backwards. Being non-deterministic means the machine can be in multiple states at once, and the machine as a whole is in an accepting state if any of those states are accepting. The initial state would be an accept state so it could recognize no suffix (as in English <code>be</code>). From the initial state, transitions <code>()</code>, <code>('e', 'z', 'i')</code>, <code>('e', 'd', 'a')</code> and <code>('e', 'v', 'o')</code> for example would all arrive at accepting states, and you don't have to worry about the conflicting <code>'e'</code>s when using an NFA. From the initial state the "characters" of each word are fed in backwards. Each time the machine lands in an accept state, the remaining part of the word is looked up in your <code>lemmaformdict</code> and all the results are kept. Then processing continues until the machine's state is null (not merely non-accepting). At that point the total choices of lemmas indicated along the way lead to the possible interpretations of that word removed from context (and that should always be a small number). Exactly how you implement an NFA will determine the performance. NFAs can be converted into DFAs once constructed, so that the machine has only one state at any given time, so that might help performance without complicating construction of the machine. On the downside, you have to work with the input on an individual character level, which for Python may cost you in performance. (But if performance is that precious, maybe you should switch to C++.)
Tags
Title
singulars
PostAcceptedAnswerId
1. This table or related slice is empty.
PostParentId
1. POBuilding a lemmatizer: speed optimization
 singulars
 PostTypePostTypeId
 PTQuestion
PostTypePostTypeId
1. PTAnswer
UserLastEditorUserId
1. USwberry
UserOwnerUserId
1. USwberry
plurals
PostLinksPostIdRelatedPostId
1. This table or related slice is empty.
PostLinksRelatedPostIdPostId
1. This table or related slice is empty.
PostsAcceptedAnswerId
1. This table or related slice is empty.
PostsParentIdCreationDate
1. This table or related slice is empty.
VotesPostIdCreationDate
1. VO
 singulars
 PostPostId
 PO
 UserUserId
 This table or related slice is empty.
 VoteTypeVoteTypeId
 VTUpMod
2. VO
 singulars
 PostPostId
 PO
 UserUserId
 This table or related slice is empty.
 VoteTypeVoteTypeId
 VTUpMod
CommentsPostId
1. CO+1: as with the first answer I will give it a try. Thanks.
 singulars
 PostPostId
 PO
 UserUserId
 USroot

Querying!

Guidance

A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.