{"responseHeader":{"status":0,"QTime":7,"params":{"q":"{!q.op=AND}id:\"99730\"","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":"/f9/7f/f97f3894512064a34e0e112dccbea96ecd93d0a7.jpg","oldid_t":"compsci 7843","setname_s":"ir_computersa","restricted_i":0,"format_t":"application/pdf","modified_tdt":"2016-05-25T00:00:00Z","file_s":"/e0/4c/e04c9e09cabaeee3704f2c140b0a03d311bdbaa1.pdf","title_t":"Page 82","ocr_t":"64 deterrrilned by the heuristics applied. We then identify several criteria and propose a performance model to evaluate the \"quality\" (performance) of a heuristic which informs our research throughout this thesis. 3.1 Backtracking is General and Essential in AI Search Backtracking search is recognized as a basic algorithmic technique in computer science (1,10,11,57,87,102,166). It is extremely useful and fundamental in solving various search problems, particularly in artificial intelligence [20,138,143,208]. 3.1.1 Framework In order to apply the backtrack method, the desired solution must be expressible as ann-tuple of variables (Lo, L1 , .•. , Ln_1) where the variable Li can be a single datum or an m-tuple vector chosen from some finite set Si of values (herem may not necessarily be equal to n ). Often the problem to be solved calls for finding one vector which maximizes (or minimizes) a criterion function, or satisfies a complex set of constraints. Those criteria, constraints, and tricks and principles for deciding which among several alternative courses of action promises to be (empirically) the most effective in order to achieve some goals, are usually called heuristics (see more details in Section 3.4.1). Varying from application domains and problem solving strategies, the goals of the backtracking technique are (listed with the least difficult first): 1. verify some tuple is a member of the solutions; 2. find the first random solution; 3. find the first solution which is of certain interest; 4. find several solutions","id":99730,"created_tdt":"2016-05-25T00:00:00Z","parent_i":99969,"_version_":1679953745530585088}]},"highlighting":{"99730":{"ocr_t":[]}}}