干涉 发表于 2025-3-25 07:03:28

http://reply.papertrans.cn/67/6619/661856/661856_21.png

Institution 发表于 2025-3-25 10:54:54

Comparing Surface Formsd as “misicien”. The Edit distance algorithm measures String orthographic distance in terms of letter deletions, insertions and substitutions. The algorithm Soundex relies instead on the sounds of words for resolving cases of orthographic error. Experiment: Comparing Edit Distance and Soundex on mis

谈判 发表于 2025-3-25 13:00:12

Exploring Corpora in corpus is shown using frequency and information content measures. Quantitative exploration of co-occurrences of words, called collocations, is shown using the point-wise mutual information and other measures. Concordancers, a tool for viewing words in their immediate contextual environment withi

商议 发表于 2025-3-25 16:18:03

Words in Sequencet. Many examples are shown to understand both the strengths and weaknesses (mostly the sparse data problem) of such models. The chapter presents how to build a word trigram model and use it in sequence prediction. Experiment: Performing orthographic error correction using language models.

欢乐东方 发表于 2025-3-25 22:44:29

http://reply.papertrans.cn/67/6619/661856/661856_25.png

不透明性 发表于 2025-3-26 01:14:07

Linguistic Rolesency parsing. This sequence of processes is sometimes referred to as the NLP stack or NLP pipeline. The chapter highlights the linguistic nature of words as well as the rules guiding the organization of words into sentences. It also suggests that analysing surface forms linguistic nature can be a fi

黄瓜 发表于 2025-3-26 04:44:51

http://reply.papertrans.cn/67/6619/661856/661856_27.png

adj忧郁的 发表于 2025-3-26 08:29:44

Relatednessrd co-occurrences in corpora to infer their relatedness. First-level relatedness (based on direct co-occurrence), second-level relatedness (based on co-occurring vectors) as well as word embeddings are discussed. Experiment: Including word relatedness in context similarity evaluation to obtain bette

genesis 发表于 2025-3-26 15:16:48

Pattern-Based Relation Extractioned approaches require explicit mentions of relations in text, the chapter discusses the interdependency between relations and text types. The chapter revisits regular expressions as possible implementations of lexical and lexico-syntactic pattern searches. The chapter also presents a semi-automatic

口诀 发表于 2025-3-26 18:09:04

From Syntax to Semanticsstructured semantic representation of sentences. Frames are used to describe all kinds of events, and each frame comes with a set of frame elements or semantic roles important in describing the events. The resource FrameNet is presented as a large resource for frames. The chapter details the in-betw
页: 1 2 [3] 4 5 6
查看完整版本: Titlebook: Natural Language Understanding in a Semantic Web Context; Caroline Barrière Textbook 2016 Springer International Publishing Switzerland 20