{"responseHeader":{"status":0,"QTime":6,"params":{"q":"{!q.op=AND}id:\"99671\"","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":"/fd/da/fddad057f0e55ec7b580a9a8e381b85feef3c05c.jpg","oldid_t":"compsci 7784","setname_s":"ir_computersa","restricted_i":0,"format_t":"application/pdf","modified_tdt":"2016-05-25T00:00:00Z","file_s":"/a9/fe/a9fede13b0ce46982b335495e23977394aaab2fb.pdf","title_t":"Page 23","ocr_t":"5 3. A high degree of modularity. In some sense this means that var1ous local \"parts\" of the constraint network can be manipulated individually without worrying too much about how they will eventually be fit together into a global \"body.\" Modularity allows a complex system to be decomposed into moreĀor-} ess independent layers and allows complex information to be represented iri terms of more-or-less independent chunks of local information. Structured partitioning for high modularity in a CLP model makes parallel computation feasible. Research on CLP has been influenced by work on mathematical logic and by work on semantic networks [126]. From mathematical logic comes the emphasis on having a well-defined semantics for the model. Also, a CLP model is quite similar to a logical model, the major difference being that in logic the models are not themselves manipulable objects in the representation, while CLP models are manipulable. From semantic network research comes the idea of parallel marker propagation and the idea that everything should be specified in terms of local connections. 1.1.2 Representation There are basically four forms of representation for a CLP model (or a CLP problem), the board representation, the constraint graph representation, the tree representation, and the assertional representation. The four forms of representation are syntactically (i.e., structurally) different but semantically (i.e., functionally) equivalent. Depending on the problem structure discussed, a different form of representation can be chosen. The board (or table) is a concise and fundamental 2-dimensional matrix form of a CLP. The graph (or network) representation encodes data in some kind of","id":99671,"created_tdt":"2016-05-25T00:00:00Z","parent_i":99969,"_version_":1679953745510662145}]},"highlighting":{"99671":{"ocr_t":[]}}}