WebOQL: Restructuring documents, databases, and webs

AutoRM: An effective approach for automatic Web data record mining

“…The algorithm for similar C-Records mining According to DEPTA [40], there may be multiple sets of similar data records under the same parent node. Thus, if T:root (the root node of T) contains child node(s), then we mine all sets of similar C-Records from T:root:children (the child nodes of T:root) (lines [3][4]. Specifically, first we cluster similar nodes in T:root:children (Note: we obtain the trees rooted at the nodes in T:root:children, and cluster similar nodes by clustering similar obtained trees using the algorithm described in Section 2.2) (line 3); second, we use Procedure mineCRecFromNS() to mine similar C-Records from T:root:children based on the similar nodes in T:root:children (line 4).…”

“…1(b), the clusters of similar non-separator nodes are {#text 3 , #text 6 , #text 14 }, {a 7 , a 15 }, {span 9 } and {font 11 , font 17 }. Thus the initial set of index intervals of these clusters are { [3,14], [7,15], [9,9], [11,17]}. We repeatedly merge two intersecting intervals in the interval set as follows:…”

“…(1) As [3,14] and [7,15] intersect, merge the two intervals into [3,14] [ [7,15] = [3,15]. The index interval set becomes { [3,15], [9,9], [11,17]}.…”

A best-effort approach to an infrastructure for Chinese Web related research

“…TGM is similar to WebSQL [10] and WebOQL [11] in that TGM accepts high-level descriptive queries as description on required services. The difference is that in TGM, attributes are used as query conditions.…”

TEX: An efficient and effective unsupervised Web information extractor

“…Although they can be handcrafted [15,24,4,42,51,50,20], the costs involved motivated many researchers to work on proposals to learn them automatically. These proposals are either supervised, i.e., they require the user to provide a number of information samples to be extracted [11,44,58,26,32,8,22,9,14,18,30,5,40,21,59], or unsupervised, i.e., they extract as much prospective information as they can and the user then gathers the relevant information from the results [62,12,16,2,28,25,60,39,46,64,67,38,59,57].…”