China
Africa
Explore chapters and articles related to this topic
The Early Shaping of Cognitive Science by Artificial Intelligence
Published in Alessio Plebe, Pietro Perconti, The Future of the Artificial Mind, 2021
Alessio Plebe, Pietro Perconti
The alternative approach mentioned by Winograd was an active area of research in Chomskyan linguistics, aimed at finding algorithms for constructing syntactic trees of a given sentence. These algorithms are called parsers, and can follow two different strategies, as sketched in Figure 3.4. Parsing always starts from a tree where the only known nodes are the root S and the leaves, made by the terminal elements of the input sentence. The bottom-up strategy proceeds shifting pairs of leaves from left to right into the parsing mechanism, then checking if the two terminal elements can be reduced in a non-terminal category by a rewriting rule of the grammar Due to this process, these parsers are also called shift-reduce. One of the earlier bottom-up algorithms is known as CKY, by the names of the three authors (Kasami, 1965). In the top-down strategy, the unknown nodes are filled by predicting which non-terminal expansions, starting from S and descending along the leftmost nodes, will match the terminal elements. For this reason, the bottom-up parsers are also called predict-match. One of the earlier algorithms in this category is thanks to Jay Earley (1970). Winograd adopted a top-down left-to-right parser.
Natural language understanding
Published in Janet Finlay, Alan Dix, An Introduction to Artificial Intelligence, 2020
Parsing may be top down, in which case it starts with the symbol for a sentence and tries to map possible rules to the input (or target) sentence, or bottom up, where it starts with the input sentence and works towards the sentence symbol, considering all the possible representations of the input sentence. The choice of which type of parsing to use is similar to that for top-down or bottom-up reasoning; it depends on factors such as the amount of branching each will require and the availability of heuristics for evaluating progress. In practice, a combination is sometimes used. There are a number of parsing methods. These include grammars, transition networks, context-sensitive grammars and augmented transition networks. As we shall see, each has its benefits and drawbacks.
L
Published in Phillip A. Laplante, Dictionary of Computer Science, Engineering, and Technology, 2017
LR(k) a bottom-up parsing algorithm that processes the input from left to right, producing a rightmost derivation (in reverse) using k tokens of lookahead. The term can also be applied to a language that can be unambiguously parsed using this algorithm. Contrast with LL(k),LALR(k).
Design and management of software development projects under rework uncertainty: a study using system dynamics
Published in Journal of Decision Systems, 2023
Mst Taskia Khatun, Kazuo Hiekata, Yutaka Takahashi, Isaac Okada
Simulation is an effective and powerful tool for analyzingdynamic behaviour and feedback responses. Two popular methods are used for dynamic simulation: SD and agent-based modelling (ABM; Alshammri & Qin, 2017, December 3-8). SD is an aggregated top-down approach that identifies dynamic effects under different conditions that affectsystem behaviour. ABM is a disaggregated bottom-up approach that defines the emergent behaviour of agents and systems (Alshammri & Qin, 2017, December 3-8).On the other hand,the agile software development process is a bottom-up approach that uses an iteration process and a short development cycle to identify different feedback conditions and reduce the risk of project failure(Cao et al., 2010). However, when considering rework in software development, a rework cycle built with SD can eradicate all rework compared to agile processes.
External validity and factor structure of individual and group workload ratings
Published in Theoretical Issues in Ergonomics Science, 2018
Stephen J. Guastello, David E. Marra
These group processes are emergent when they acquire a stable pattern that persists over time (Goldstein 2011; Guastello and Liebovitch 2009; Sawyer 2005), at which point there is a ‘whole that is greater than the sum of its parts’. This is a bottom-up process wherein interactions among the individuals give rise to a global pattern of group interaction. According to Goldstein (2011), agents within a system interact according to unspecified rules, which can develop extemporaneously, and the rules contain biases that eventually lead to particular group structures. Interaction patterns would include who talks to whom and in what order, how the participants facilitate each other's ideas, and subtexts such as friendliness, hostility, closeness and control of the task or conversation, and expectations of other team members’ competence (Bales 1999; Balkwell 1991; Correll and Ridgeway 2006; Guastello 2010; Pincus and Guastello 2005). Biases can result from any of a number of cognitive or personality traits that were formed prior to the group interaction. In dyadic or team coordination, bottom-up effects arise when individuals develop similar perception-action patterns (Gipson, Gorman, and Hessler 2016).
Eye-Tracking Studies Based on Attentional-Resource Effectiveness and Insights into Future Research
Published in Nuclear Technology, 2018
Jun Su Ha, Young-Ji Byon, Chung-Suk Cho, Poong Hyun Seong
Operators in NPPs continuously monitor the status of the plant system during normal operation. When they detect symptomatic events representing abnormal situations, they actively search for relevant information to correctly understand the situation. Stages of information processing depend on mental or cognitive resources, a sort of pool of attention or mental effort that is of limited availability and can be allocated to processes as required.5 With regard to attentional resources, there are two aspects of attention: selecting information sources for further information processing and dividing attention between tasks. Attention is typically driven by four factors: salience, expectancy, value, and effort.6 Salience refers to stimuli in the environment such as alarms, alerts, or some remarkable indication representing deviation from the normal situation. Expectancy shifts attention to specific sources that are most likely to provide information. Frequency of looking at or attending to an information source is modified by how valuable it is to look at. Attention may be inhibited if it is effortful compared to its value. Perception or understanding is accomplished by three simultaneous processes: bottom-up processing, top-down processing, and unitization (or matching) of the two processes. Bottom-up processing is derived by stimulus or salient information sources through sensing mechanisms. After detecting a stimulus, the information is matched to a mental model that is established based on knowledge and experience. Expectancy derived from the mental model leads to effective selection of information sources, which is top-down processing. The series of bottom-up processing, top-down processing, and unitization is the process of perception or understanding.