BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:women@CL Talklets - NLIP - Computer Laboratory\, University of Cam
 bridge
DTSTART:20150212T130000Z
DTEND:20150212T140000Z
UID:TALK55021@talks.cam.ac.uk
CONTACT:Ekaterina Kochmar
DESCRIPTION:Title: Learning from Text Whether a Lion is an Animal\n\nSpeak
 er: Laura Rimell\n\nAbstract: The ability to detect entailment relations b
 etween natural language sentences is a crucial Artificial Intelligence tas
 k. A prerequisite is the ability to automatically identify entailment rela
 tions between words: for example\, being a lion entails being an animal\, 
 and kicking something entails touching it. This talk will discuss how such
  relations can be learned in an unsupervised way from large amounts of tex
 t. We make use of distributional semantics\, the idea that a word's contex
 ts of use characterise its meaning. The talk will begin with an introducti
 on to distributional semantics and describe how the contexts of words like
  "lion" and "animal" relate to each other in surprising ways.\n\n\nTitle: 
 Words as Functions: Composing Meanings\n\nSpeaker: Tamara Polajnar\n\nAbst
 ract: Although we are able to adequately\, if not perfectly\, represent th
 e meaning of words in a way that a computer can process\, longer construct
 ions such as phrases and sentences offer new challenges. In this talk\, I 
 will describe how we take into account grammatical structure in order to e
 ncode the meaning of simple sentences in a vector space. There are many ta
 sks where the word order\, grammatical structure\, and semantics are key. 
 For example\, we may want to be able to distinguish that in a sentence lik
 e "Animals eat plants"\, the animals are the animate subjects with teeth d
 oing the eating and not the other way around. Likewise\, we may be concern
 ed about plausibility and want to reject sentences like "Animals eat plane
 ts".\n\n\nTitle: Error Detection in Non-Native English Writing\n\nSpeaker:
  Ekaterina Kochmar\n\nAbstract: Even though English is used as the interna
 tional language in many fields\, the majority of people who use it are not
  native speakers of English. As a result\, they inevitably commit errors. 
 Errors in the use of content words (for example\, adjectives\, nouns and v
 erbs) can change the meaning of the original expression or even result in 
 implausible word combinations. For example\, if non-native speakers write 
 "Animals eat planets" they probably mean "plants“\, and when they write 
 "big history“ they ptobably mean "long history“.\nThis talk will discu
 ss error detection and correction in the use of content words by non-nativ
 e speakers of English. We model the meaning of the words computationally u
 sing the methods of compostional and distributional semantics overviewed i
 n the previous talks. The implemented error detection and correction algor
 ithm can further be integrated in a self-assessment and tutoring system.\n
 \n\nTitle: Write & Improve!\n\nSpeaker: Helen Yannakoudakis\n\nAbstract: T
 he task of automated assessment (AA) of text focuses on automatically anal
 ysing and assessing the quality of writing. A number of AA systems have be
 en developed\, aimed particularly at second-language learners and providin
 g feedback on their writing. Automated writing feedback may be a useful co
 mplement to teacher comments in the process of learning a foreign language
 \, while recent studies have shown that automated writing evaluation can l
 ead to increased learner autonomy and higher writing accuracy. In this tal
 k\, we will discuss how we can leverage natural language processing and ma
 chine learning techniques to develop an accurate self-assessment and tutor
 ing system efficiently. We will describe Write & Improve\, a system that p
 rovides automated feedback on learners' writing at three different levels 
 of granularity: 1) an overall assessment of their proficiency\, 2) a quali
 tative assessment of each individual sentence\, making the language learne
 r aware of potentially problematic areas rather than providing a panacea\,
  and 3) diagnostic feedback on local issues including spelling and word ch
 oice. Pedagogically useful feedback requires high precision (few false pos
 itives in the detection of errors) and reasonable recall (not too many err
 ors undetected)\, which makes this a challenging task.\n
LOCATION:Computer Laboratory\, William Gates Building\, Room FW26
END:VEVENT
END:VCALENDAR