Implicatives & factives [Nairn et al. 06]

فهرست عناوین اصلی در این پاورپوینت

فهرست عناوین اصلی در این پاورپوینت

● Natural Language Inference
● Natural language inference (NLI)
● Applications of NLI
● NLI problem sets
● NLI: a spectrum of approaches
● Shallow approaches to NLI
● The formal approach to NLI
● Solution? Natural logic! ( natural deduction)
● Outline
● Alignment for NLI
● Alignment example
● Approaches to NLI alignment
● The MANLI aligner
● Phrase-based alignment representation
● A feature-based scoring function
● Decoding using simulated annealing
● Perceptron learning of feature weights
● The MSR RTE2 alignment data
● Evaluation on MSR data
● Aligner evaluation results
● MANLI results: discussion
● Alignment for NLI: conclusions
● Outline
● Entailment relations in past work
● 16 elementary set relations
● The set of basic entailment relations
● Joining entailment relations
● Some joins yield unions of relations!
● The complete join table
● Outline
● Lexical entailment relations
● Lexical entailment relations: SUBs
● Lexical entailment relations: DEL & INS
● The impact of semantic composition
● A typology of projectivity
● Projecting through multiple levels
● Implicatives & factives [Nairn et al. 06] ● Implicatives & factives
● Putting it all together
● An example
● Different edit orders?
● Outline
● The NatLog system
● Stage 1: Linguistic analysis
● Stage 3: Lexical entailment classification
● The FraCaS test suite
● Results on FraCaS
● The RTE3 test suite
● Results on RTE3: NatLog
● Results on RTE3: hybrid system
● Outline
● What natural logic can’t do
● What natural logic can do
● Contributions of this dissertation
● The future of NLI
● Thanks!
● Backup slides follow
● NLI alignment vs. MT alignment
● Projectivity of connectives
● Projectivity of quantifiers

نوع زبان: انگلیسی حجم: 2.38 مگا بایت
نوع فایل: اسلاید پاورپوینت تعداد اسلایدها: 67 صفحه
سطح مطلب: نامشخص پسوند فایل: ppt
گروه موضوعی: زمان استخراج مطلب: 2019/05/17 04:54:25

لینک دانلود رایگان لینک دانلود کمکی

اسلایدهای پاورپوینت مرتبط در پایین صفحه

عبارات مهم استفاده شده در این مطلب

عبارات مهم استفاده شده در این مطلب

., entailment, problem, inference, natural, ‘s, nlus, premise, language, gas, system, p,

توجه: این مطلب در تاریخ 2019/05/17 04:54:25 به صورت خودکار از فضای وب آشکار توسط موتور جستجوی پاورپوینت جمع آوری شده است و در صورت اعلام عدم رضایت تهیه کننده ی آن، طبق قوانین سایت از روی وب گاه حذف خواهد شد. این مطلب از وب سایت زیر استخراج شده است و مسئولیت انتشار آن با منبع اصلی است.

https://nlp.stanford.edu/~wcmac/papers/nli-diss.ppt

در صورتی که محتوای فایل ارائه شده با عنوان مطلب سازگار نبود یا مطلب مذکور خلاف قوانین کشور بود لطفا در بخش دیدگاه (در پایین صفحه) به ما اطلاع دهید تا بعد از بررسی در کوتاه ترین زمان نسبت به حدف با اصلاح آن اقدام نماییم. جهت جستجوی پاورپوینت های بیشتر بر روی اینجا کلیک کنید.

عبارات پرتکرار و مهم در این اسلاید عبارتند از: ., entailment, problem, inference, natural, ‘s, nlus, premise, language, gas, system, p,

مشاهده محتوای متنیِ این اسلاید ppt

مشاهده محتوای متنیِ این اسلاید ppt

natural language inference bill maccartney nlp group stanford university ۸ may ۲ ۹ thank you all very much for coming today. i’m happy to have the opportunity to tell you about my work. so let’s dive right in natural language inference nli aka recognizing textual entailment rte does premise p justify an inference to hypothesis h an informal intuitive notion of inference not strict logic emphasis on variability of linguistic expression necessary to goal of natural language understanding nlu many more immediate applications … p several airlines polled saw costs grow more than expected even after adjusting for inflation. h some of the companies in the poll reported cost increases. yes introduction alignment for nli entailment relations compositional entailment the natlog system conclusion natural language inference also known as recognizing textual entailment is the problem of determining whether a premise p justifies an inference to a hypothesis h. this is an informal intuitive notion of inference the emphasis is on short local inference steps and variability of linguistic expression rather than on long chains of formal reasoning. here’s an example. the premise is … while the hypothesis is … and this a valid inference. but note that h is not a strict logical consequence of p for one thing seeing cost increases does not necessarily entail reporting cost increases. and this underscores the fact that this task relies on an informal standard of inferability given the premise what would most ordinary people accept as a valid conclusion a capacity for natural language inference is a necessary condition for achieving the ultimate goal of full natural language understanding but it can also enable more immediate applications. applications of nli input gazprom va doubler le prix du gaz pour la géorgie. semantic search question answering summarization mt evaluation x x output gazprom will double the price of gas for georgia. target gazprom will double georgia’s gas bill. machine translation evaluation does output paraphrase target georgia’s gas bill doubled search q how much did georgia’s gas price increase a in ۲ ۶ gazprom doubled georgia’s gas bill. a georgia’s main imports are natural gas machinery … a tbilisi is the capital and largest city of georgia. a natural gas is a gas consisting primarily of methane. introduction alignment for nli entailment relations compositional entailment the natlog system conclusion pado et al. ۹ harabagiu hickl ۶ tatar et al. ۸ king et al. ۷ here’s a quick sketch of four potential applications. one is semantic search if the text of a web page entails the search query then it’s probably a relevant result. likewise in question answering if a candidate answer entails the question — or a declarative form of it — then it’s probably a good answer. in summarization the key challenge is eliminating redundancies thus we need the ability to recognize when one sentence is a paraphrase of another. identifying paraphrases amounts to identifying approximate bidirectional entailment so it’s a special case of natural language inference. finally nli can help us to automatically evaluate the quality of a machine translation system by checking whether the system output is a good paraphrase of a reference translation. nli problem sets rte recognizing textual entailment ۴ years each with dev test sets each ۸ nli problems longish premises taken from e.g. newswire short hypotheses balanced ۲ way classification entailment vs. non entailment introduction alignment for nli entailment relations compositional entailment the natlog system conclusion fracas test suite ۳۴۶ nli problems constructed by semanticists in mid ۹ s ۵۵ have single premise remainder have ۲ or more premises ۳ way classification entailment contradiction compatibility p as leaders gather in argentina ahead of this weekends regional talks hugo chávez venezuela’s populist president is using an energy windfall to win friends and promote his vision of ۲۱st century socialism. h hugo chávez acts as venezuela’s president. yes p smith wrote a report in two hours. h smith spend more than two hours writing the report. no nli systems are developed and evaluated using standard collections of nli problems and today we’re going to look at two such collections. the first comes from the well known rte or recognizing textual entailment challenge which has been held every year for the last four years. the rte problems have fairly long premises typically harvested from newswire text and use fairly short hypotheses. and they cover a broad range of types of inference. the rte task is a binary classification task entailment vs. non entailment and the problem sets are balanced between yes and no answers. the other problem set we’ll look at is the fracas test suite. this is a much smaller collection of problems with shorter premises and the problems tend to hinge on well known semantic phenomena. fracas uses three way classification it divides non entailment into contradiction and compatibility. and i’ve shown a couple of example problems here. nli a spectrum of approaches problem imprecise  easily confounded by negation quantifiers conditionals factive implicative verbs etc. problem hard to translate nl to fol idioms anaphora ellipsis intensionality tense aspect vagueness modals indexicals reciprocals propositional attitudes scope ambiguities anaphoric adjectives non intersective adjectives temporal causal relations unselective quantifiers adverbs of quantification donkey sentences generic determiners comparatives phrasal verbs … solution introduction alignment for nli entailment relations compositional entailment the natlog system conclusion work on natural language inference has explored a broad spectrum of approaches. at one end of the spectrum are shallow approaches …

کلمات کلیدی پرکاربرد در این اسلاید پاورپوینت: ., entailment, problem, inference, natural, ‘s, nlus, premise, language, gas, system, p,

این فایل پاورپوینت شامل 67 اسلاید و به زبان انگلیسی و حجم آن 2.38 مگا بایت است. نوع قالب فایل ppt بوده که با این لینک قابل دانلود است. این مطلب برگرفته از سایت زیر است و مسئولیت انتشار آن با منبع اصلی می باشد که در تاریخ 2019/05/17 04:54:25 استخراج شده است.

https://nlp.stanford.edu/~wcmac/papers/nli-diss.ppt

  • جهت آموزش های پاورپوینت بر روی اینجا کلیک کنید.
  • جهت دانلود رایگان قالب های حرفه ای پاورپوینت بر روی اینجا کلیک کنید.

رفتن به مشاهده اسلاید در بالای صفحه


پاسخی بگذارید

نشانی ایمیل شما منتشر نخواهد شد. بخش‌های موردنیاز علامت‌گذاری شده‌اند *