Recognizing Instantiated Goals using Statistical Methods
نویسندگان
چکیده
We present our work on using statistical, corpusbased machine learning techniques to perform instantiated goal recognition — recognition of a goal schema and its parameter values. The recognizer is fast (linear in the number of goal schemas and observed actions) and is able to make partial predictions by optionally predicting individual parameter values for a goal schema. This allows it to make more accurate predictions after observing fewer actions than all-or-nothing prediction.
منابع مشابه
Recognizing Plan/Goal Abandonment
The ability to recognize when an agent abandons a plan is an open problem in the plan recognition literature and is a significant problem if these methods are to be applied in real systems. This paper presents an explicit, formal, and implemented solution to the problem of recognizing when an agent has abandoned one of its goals based on a theory of probabilistic model revision.
متن کاملStudies on Automatic Recognition of Common Chinese Adverb's usages Based on Statistics Methods
The study on Automatic Recognizing usages of Modern Chinese Adverbs is one of the important parts of the NLP-oriented research of Chinese Functional words Knowledge Base. To solve the problems of the existing rule-based method of adverbs’ usages recognition based on the previous work, this paper has studied automatically recognizing common Chinese adverbs’ usages using statistical methods. Thre...
متن کاملAssessing the information literacy of Farhangian University student-teachers from six perspectives: reviewing, recognizing sources, disseminating, recognizing, flexibility and seeking information
Background: Today, the scope of science and knowledge and, consequently, its measurement has become very common. The use of this information in the university environment depends on the students' knowledge of the places where the information is disseminated and how it is used and the methods of retrieving and using that information. The aim of this study was to investigate the information liter...
متن کاملTowards an Assembly Plan from Observation: Task recognition with polyhedral objects
task models instantiated task model I Grasp Recognition recognizing where and how the human operator grasps an Object for achieving the assembly
متن کاملRepresentation and Visual Recognition of Complex , Multi - agent Actions using Belief
A probabilistic framework for representing and visually recognizing complex multi-agent action is presented. Motivated by work in model-based object recognition and designed for the recognition of action from visual evidence , the representation has three components: (1) temporal structure descriptions representing the logical and temporal relationships between agent goals, (2) belief networks ...
متن کامل