{"responseHeader":{"status":0,"QTime":6,"params":{"q":"{!q.op=AND}id:\"99745\"","hl":"true","hl.simple.post":"","hl.fragsize":"5000","fq":"!embargo_tdt:[NOW TO *]","hl.fl":"ocr_t","hl.method":"unified","wt":"json","hl.simple.pre":""}},"response":{"numFound":1,"start":0,"docs":[{"file_name_t":"Gu-Parallel_Algorithms.pdf","thumb_s":"/5e/e6/5ee62576f7ab625a5752ffada596c73d899df356.jpg","oldid_t":"compsci 7858","setname_s":"ir_computersa","restricted_i":0,"format_t":"application/pdf","modified_tdt":"2016-05-25T00:00:00Z","file_s":"/c3/e6/c3e64c2d401bbf4ff223764796e36daa471acf9b.pdf","title_t":"Page 97","ocr_t":"Each of the future nodes in a CLP search tree determines a future search state, i.e., a future partial path and a future search space. 79 In a CLP search toward a complete solution path, the current search state is built by • concatenating the current node to last partial path, and • pruning those nodes whose label values are inconsistent with the current node. Each of the successive future search states is built in a similar way. Note that a search state is defined for \"a node\" (actually an \"object-value\" pair) and each search state of a tree node can be represented by using a CLP board. Each CLP board of search state remembers a historical consequence of an accumulated partial path and a graduatedly reduced search space 3.2.2 Basic CLP Algorithm Intuitively, a CLP search visits each and every label on a problem board in a linear ordering, temporarily assigning consistent values to a sublist (![row OJ, l[row 1], ... , ![row i]) of labels and attempting to concatenate a new label instance ![row i + 1] from the ( i + 1 )th row on the board such that this set of labels is consistent. Backtracking occurs if this new label is not consistent with the previous set of labels already found. A solution of the CLP, i.e., a consistent set of labels, is found if and only if this set of labels contains n consistent labels and each of them cames from a separate row on a CLP board. A basic, recursive CLP algorithm embedding the above ideas is given in Figure 3.3. The algorithm is invoked by a call to Primi t i ve_CLP ( 0) and is terminated after walking through the bottom of the tree. It is able to find all solutions of a CLP problem. This algorithm consists of mainly two alternative procedures:","id":99745,"created_tdt":"2016-05-25T00:00:00Z","parent_i":99969,"_version_":1679953745534779394}]},"highlighting":{"99745":{"ocr_t":[]}}}