"... In PAGE 6: ...Table 1. Classification of AI, OR and MS techniques for ISE Table1 shows where each technique can be used in relation to a-priori or a-posteriori ISE: the former in the sense of allowing the research to identify factors to be used within the ISE decision-making process; and the latter in terms of the ability to predict and interpret decisions, after the ISE process has taken place occurred. Each of the investigated approaches has been categorised in terms of four key forms of modelling approach: Cognitive (based upon mental or conceptual models of the process and / organisational inter-relationships); Systematic (based upon an explicit representation of knowledge using measureable data and information); Non-Deterministic (using techniques which do not assume or require convergence to a pre-defined goal state); Stochastic (based upon variations of statistically distributed, random variables and their associated outcomes relative to historical behaviour).... ..."

"... In PAGE 4: ... There have been a number of attempts to combine techniques available at a given time into prototypes able to cope with increasingly more realistic application domain. Table1 lists some of these efforts and the domains to... ..."

"... In PAGE 25: ... To maximize lnP (DjN), we need to know the distribution that the network apos;s outputs are to learn. Table1 shows three probability distributions that are commonly assumed in practice, and their appropriate error functions (Rumelhart et al., 1995).... In PAGE 88: ... Table1 0: Test-set error from a ten-fold cross validation. Table (a) shows the results from running three learners without the domain theory; Table (b) shows the results of running three learners with the domain theory.... In PAGE 88: ...5% 34.7% (b) Generating Non-KNN Ensembles Table1 0a presents the results from the case where the learners randomly create the topology of their networks (i.... In PAGE 88: ...opology of their networks (i.e., they do not use the domain theory). Table1 0a apos;s rst row, best-network, results from a single-layer neural network where, for each fold, I trained 20 networks containing between 0 and 100 (uniformly) hidden nodes and used a validation set to choose the best network. The next row, bagging, contains the results of applying the bagging algorithm to standard, single-hidden-layer networks, where the number of hidden nodes is randomly set between 0 and 100 for each network.... In PAGE 89: ...can produce large alterations in the predictions of the networks, thereby leading to an e ective ensemble. The bottom row of Table1 0a, Addemup, contains the results of a run of Addemup where its initial population (of size 20) is randomly generated using Regent apos;s method for creating networks when no domain theory is present (refer to Section 4.3.... In PAGE 89: ... Again, each ensemble con- tains 20 networks. The rst row of Table1 0b contains the generalization results of the Kbann algorithm, while the next row, Kbann-bagging, contains the results of the en- semble where each individual network in the ensemble is the Kbann network trained on a di erent partition of the training set. Even though each of these networks start with the same topology and \large quot; initial weight settings (i.... In PAGE 89: ...e., bagging in Table1 0a). The next two rows result from the Regent algorithm.... In PAGE 89: ... The rst row, Regent-best- network, contains the results from the single best network output by Regent, while the next row, Regent-combined, contains the results of simply combining, using Equa- tion 21, the networks in Regent apos;s nal population. Last chapter, I showed the e ective- ness of Regent-best-network, and comparing it with the results in Table1... In PAGE 90: ... Notice that simply combining the networks of Regent apos;s nal population (Regent-combined) decreases the test-set error over the single-best network picked by Regent. Addemup, the nal row of Table1 0b, mainly di ers from Regent-combined in two ways: (a) its tness function (i.e.... In PAGE 90: ...e., a repeat of Addemup from Table1 0b). The results show that, while reweighting the examples during training usually helps, Addemup gets most of its generalization power from its tness function.... In PAGE 91: ... Table1 1: Test-set error on the lesion studies of Addemup. Due to the inherent similarity of each algorithm and the lengthy run-times limiting the number of runs to a ten-fold cross-validation, the di erence between the lesions of Addemup is not signi cant at the 95% con dence level.... In PAGE 92: ...show that more diversity is needed when generating an e ective ensemble. There are two main reasons why I think the results of Addemup in Table1 0b are especially encouraging: (a) by comparing Addemup with Regent-combined, I explicitly test the quality of my tness function and demonstrate its e ectiveness, and (b) Addemup is able to e ectively utilize background knowledge to decrease the error of the individual networks in its ensemble, while still being able to create enough diversity among them so as to improve the overall quality of the ensemble. 5.... In PAGE 119: ... Table1 2: Domain theory for the arti cial chess problem. The OR in the following rules means that only one of the two antecedents needs to be satis ed in order for the rule to be satis ed.... In PAGE 124: ... Table1 3: Domain theory for nding promoters.... In PAGE 128: ... Table1 5: Domain theory for nding ribosome-binding sites.... In PAGE 130: ... Table1 6: Part 1 of the domain theory for nding transcription-termination sites; see Table 17 for the remainder of this theory.... In PAGE 131: ... Table1 7: Part 2 of the domain theory for nding transcription-termination sites; Table 16 contains the rst half of this theory. % A region upstream from the site should be rich with A apos;s and T apos;s.... In PAGE 131: ... loop-even :- @-2= quot;TTTT quot;. % As in Table1 6, this rule encodes the notion that a longer stem is more stable. Like % before, the consequent of this rule, stem-even, is true if the sum of the \weighted quot; % true antecedents is greater than 6.... ..."

"... In PAGE 49: ...Measures Sprincles and Fire Distincuters Gravity Very High Consequences Complete Destruction of the Restaurent Potential Accidents Fire Events Causing Potential Accidents No Extincture Dangerous Situation Beginning of fire Events Causing Dangerous Situation Contact between oil and heat source Dangerous Element Oil Oven Heaters Phase Food Preparation System or Sub-System Restaurants Table1... In PAGE 67: ...Random Random Observation Number ri xi=-ln(1-ri) 1 0,9609 3,2412 2 0,0400 0,0409 3 0,3689 0,4603 4 0,5164 0,7265 5 0,6563 1,0679 6 0,2843 0,3344 7 0,9063 2,3680 8 0,3298 0,4002 9 0,0545 0,0560 10 0,3296 0,3999 Total 9,0952 Mean 0,9095 Table1 . Random Observations Determination of the critical index of activities in a network could be achieved with the application of simulation (Monte Carlo).... In PAGE 71: ... Different ports or phases have their own standard notation and have to be read and understood correctly in order to proceed to the analysis of the tree, these ports or operations are classified to Fundamental Operations, Special Operations, Events, and Transfer Triangles. Table1... In PAGE 72: ...Name Explaination OR The Exit is generated if one of the entries exist AND The exit is generated if all the entries exist Table1... In PAGE 80: ...amples (100 samples). Table 2. act. Min Max s1 s2 s3 s4 s5 s6 s7 s8 s9 s10 Mean STD 1-2 8 10 9,5 8,6 8,1 8,3 8,4 8,9 8,0 9,5 9,7 8,5 8,8 0,60 1-3 5 7 6,3 5,1 7,0 5,5 6,4 5,2 5,7 6,5 6,5 5,0 5,9 0,71 1-4 7 9 8,4 8,1 8,3 7,7 8,2 7,5 7,4 7,1 7,5 7,5 7,8 0,45 2-5 5 7 6,7 5,1 5,3 5,7 6,9 5,3 6,1 5,6 5,9 5,8 5,9 0,58 3-6 4 6 4,9 4,2 4,3 5,7 5,5 5,1 5,3 5,4 5,3 4,3 5,0 0,54 3-5 8 10 9,0 9,9 9,9 9,1 8,4 9,3 8,7 8,3 9,1 9,8 9,2 0,60 3-7 11 13 11,4 12,8 12,9 11,6 12,9 12,5 12,8 12,6 11,5 11,6 12,3 0,64 4-7 10 12 10,8 11,3 10,2 10,2 10,8 11,1 10,2 10,3 11,2 11,8 10,8 0,57 5-8 6 8 6,1 6,4 7,1 6,5 6,4 7,0 6,9 7,0 7,9 6,8 6,8 0,50 5-9 6 8 7,4 6,4 7,1 7,8 8,0 6,9 6,7 6,5 6,1 6,3 6,9 0,65 6-8 2 4 2,4 2,1 2,5 2,3 2,8 3,3 2,9 2,4 2,5 2,6 2,6 0,36 7-8 0 2 1,9 0,5 1,6 1,7 0,5 0,4 0,6 1,2 0,2 1,0 1,0 0,61 7-10 3 5 4,6 3,5 3,7 3,7 4,8 4,0 4,2 4,3 3,0 3,1 3,9 0,60 8-10 2 4 2,6 3,6 2,2 3,9 3,6 2,5 2,2 3,0 2,2 3,5 2,9 0,69 9-10 0 3 0,3 0,6 2,7 2,2 2,5 2,8 1,5 1,2 0,7 1,8 1,6 0,92 Total 72 96 82,8 Table1 , Random Observations act. Min Max s1 s2 s3 s4 s5 s6 s7 s8 s9 s10 Critical Index 1-2 8 10 * * * * 40% 1-3 5 7 * * * * * * * 70% 1-4 7 9 0% 2-5 5 7 * * * * 40% 3-6 4 6 0% 3-5 8 10 * * * * * * * 70% 3-7 11 13 0% 4-7 10 12 0% 5-8 6 8 * * * * * * * 70% 5-9 6 8 * * * 30% 6-8 2 4 0% 7-8 0 2 0% 7-10 3 5 0% 8-10 2 4 * * * * * * * 70% 9-10 0 3 * * * 30% Table 2, Critical Index of activities (activities with * means that it was on the Critical Path in this sample).... ..."

"... In PAGE 9: ... [104] also does an excellent job of distinguishing \traditional AI quot; from the study of autonomous agents (what we apos;ll term agent AI). Table1 summarizes the main di erences traditional AI and agent AI. [104] discusses the basic problems AI researchers have with adaptive autonomous agents, which includes interface agents.... ..."

"... In PAGE 19: ....2.2. Planning Representations UCPOP provides its own representations for planning operators; in addition, we represented domain knowledge in structures, lists and procedures in Lisp to support the pre- and post-processors. Table7 lists each domain model component and its planning representation; each one is described subsequently. Script Representation A script class is represented as the planner apos;s domain.... ..."

"... In PAGE 10: ... This was due to the fact that, when the cell is triggered, the signal generated by the Hodgkin-Huxley model is nonlinear, as we can see in Fig- ure 12.Thus, between 0 and 2 msec were needed to approx- imate the action potential using four different polynomials (as shown in Table1 ). We also introduced an intermediate state in which the polynomial evaluation would result in obtaining a positive value, which will trigger activity in the neighboring cells in this example (polynomial P2isin charge of this).... ..."

"... In PAGE 6: ... Location of the user and the means of connectivity also have an impact on the cost and complexity of deploying an application. Table1 summarizes the application deployment characteristics for four computing models introduced in Section 1 [9]. Table 1.... ..."

"... In PAGE 5: ... The ability to invoke AI components flexibly and dynamically from within the workflow framework would considerably enhance business productivity. Table1 outlines at a high level the concepts that AI P amp;S share with the Workflow community (for a more detailed description we refer to the Workflow Man- agement PLANET TCU PLANET roadmap). Each BPR domain (e.... ..."