text
stringlengths
49
577
label
stringclasses
7 values
metadata
list
With a [[ sentence-aligned corpus ]] , << translation equivalences >> are suggested by analysing the frequency profiles of parallel concordances .
USED-FOR
[ 2, 3, 5, 6 ]
With a sentence-aligned corpus , translation equivalences are suggested by analysing the [[ frequency profiles ]] of << parallel concordances >> .
PART-OF
[ 12, 13, 15, 16 ]
The [[ method ]] overcomes the limitations of conventional << statistical methods >> which require large corpora to be effective , and lexical approaches which depend on existing bilingual dictionaries .
COMPARE
[ 1, 1, 7, 8 ]
The [[ method ]] overcomes the limitations of conventional statistical methods which require large corpora to be effective , and << lexical approaches >> which depend on existing bilingual dictionaries .
COMPARE
[ 1, 1, 18, 19 ]
The method overcomes the limitations of conventional << statistical methods >> which require [[ large corpora ]] to be effective , and lexical approaches which depend on existing bilingual dictionaries .
USED-FOR
[ 11, 12, 7, 8 ]
The method overcomes the limitations of conventional statistical methods which require large corpora to be effective , and << lexical approaches >> which depend on existing [[ bilingual dictionaries ]] .
USED-FOR
[ 24, 25, 18, 19 ]
Pilot testing on a parallel corpus of about 113K Chinese words and 120K English words gives an encouraging 85 % [[ precision ]] and 45 % << recall >> .
CONJUNCTION
[ 20, 20, 24, 24 ]
Future work includes fine-tuning the algorithm upon the analysis of the errors , and acquiring a [[ translation lexicon ]] for << legal terminology >> by filtering out general terms .
USED-FOR
[ 16, 17, 19, 20 ]
Traditional [[ machine learning techniques ]] have been applied to this << problem >> with reasonable success , but they have been shown to work well only when there is a good match between the training and test data with respect to topic .
USED-FOR
[ 1, 3, 9, 9 ]
This paper demonstrates that match with respect to domain and time is also important , and presents preliminary experiments with << training data >> labeled with [[ emoticons ]] , which has the potential of being independent of domain , topic and time .
FEATURE-OF
[ 24, 24, 20, 21 ]
We present a novel [[ algorithm ]] for estimating the broad << 3D geometric structure of outdoor video scenes >> .
USED-FOR
[ 4, 4, 9, 15 ]
Leveraging [[ spatio-temporal video segmentation ]] , we decompose a << dynamic scene >> captured by a video into geometric classes , based on predictions made by region-classifiers that are trained on appearance and motion features .
USED-FOR
[ 1, 3, 8, 9 ]
Leveraging spatio-temporal video segmentation , we decompose a << dynamic scene >> captured by a video into [[ geometric classes ]] , based on predictions made by region-classifiers that are trained on appearance and motion features .
PART-OF
[ 15, 16, 8, 9 ]
Leveraging spatio-temporal video segmentation , we decompose a dynamic scene captured by a video into << geometric classes >> , based on predictions made by [[ region-classifiers ]] that are trained on appearance and motion features .
USED-FOR
[ 23, 23, 15, 16 ]
Leveraging spatio-temporal video segmentation , we decompose a dynamic scene captured by a video into geometric classes , based on predictions made by << region-classifiers >> that are trained on [[ appearance and motion features ]] .
USED-FOR
[ 28, 31, 23, 23 ]
We built a novel , extensive [[ dataset ]] on geometric context of video to evaluate our << method >> , consisting of over 100 ground-truth annotated outdoor videos with over 20,000 frames .
EVALUATE-FOR
[ 6, 6, 15, 15 ]
We built a novel , extensive << dataset >> on [[ geometric context of video ]] to evaluate our method , consisting of over 100 ground-truth annotated outdoor videos with over 20,000 frames .
FEATURE-OF
[ 8, 11, 6, 6 ]
We built a novel , extensive << dataset >> on geometric context of video to evaluate our method , consisting of over 100 ground-truth [[ annotated outdoor videos ]] with over 20,000 frames .
PART-OF
[ 22, 24, 6, 6 ]
To further scale beyond this dataset , we propose a [[ semi-supervised learning framework ]] to expand the pool of << labeled data >> with high confidence predictions obtained from unlabeled data .
USED-FOR
[ 10, 12, 18, 19 ]
To further scale beyond this dataset , we propose a << semi-supervised learning framework >> to expand the pool of labeled data with [[ high confidence predictions ]] obtained from unlabeled data .
USED-FOR
[ 21, 23, 10, 12 ]
To further scale beyond this dataset , we propose a semi-supervised learning framework to expand the pool of labeled data with << high confidence predictions >> obtained from [[ unlabeled data ]] .
USED-FOR
[ 26, 27, 21, 23 ]
Our [[ system ]] produces an accurate prediction of << geometric context of video >> achieving 96 % accuracy across main geometric classes .
USED-FOR
[ 1, 1, 7, 10 ]
Our << system >> produces an accurate prediction of geometric context of video achieving 96 % [[ accuracy ]] across main geometric classes .
EVALUATE-FOR
[ 14, 14, 1, 1 ]
This paper describes a [[ system ]] -LRB- RAREAS -RRB- which synthesizes << marine weather forecasts >> directly from formatted weather data .
USED-FOR
[ 4, 4, 10, 12 ]
This paper describes a << system >> -LRB- RAREAS -RRB- which synthesizes marine weather forecasts directly from [[ formatted weather data ]] .
USED-FOR
[ 15, 17, 4, 4 ]
Such << synthesis >> appears feasible in certain [[ natural sublanguages with stereotyped text structure ]] .
USED-FOR
[ 6, 11, 1, 1 ]
<< RAREAS >> draws on several kinds of [[ linguistic and non-linguistic knowledge ]] and mirrors a forecaster 's apparent tendency to ascribe less precise temporal adverbs to more remote meteorological events .
USED-FOR
[ 6, 9, 0, 0 ]
The << approach >> can easily be adapted to synthesize [[ bilingual or multi-lingual texts ]] .
USED-FOR
[ 8, 11, 1, 1 ]
We go , on to describe [[ FlexP ]] , a << bottom-up pattern-matching parser >> that we have designed and implemented to provide these flexibilities for restricted natural language input to a limited-domain computer system .
HYPONYM-OF
[ 6, 6, 9, 11 ]
We go , on to describe FlexP , a [[ bottom-up pattern-matching parser ]] that we have designed and implemented to provide these << flexibilities >> for restricted natural language input to a limited-domain computer system .
USED-FOR
[ 9, 11, 21, 21 ]
We go , on to describe FlexP , a bottom-up pattern-matching parser that we have designed and implemented to provide these [[ flexibilities ]] for << restricted natural language >> input to a limited-domain computer system .
FEATURE-OF
[ 21, 21, 23, 25 ]
We go , on to describe FlexP , a bottom-up pattern-matching parser that we have designed and implemented to provide these [[ flexibilities ]] for restricted natural language input to a << limited-domain computer system >> .
PART-OF
[ 21, 21, 29, 31 ]
We go , on to describe FlexP , a << bottom-up pattern-matching parser >> that we have designed and implemented to provide these flexibilities for [[ restricted natural language ]] input to a limited-domain computer system .
USED-FOR
[ 23, 25, 9, 11 ]
Traditional << information retrieval techniques >> use a [[ histogram of keywords ]] as the document representation but oral communication may offer additional indices such as the time and place of the rejoinder and the attendance .
USED-FOR
[ 6, 8, 1, 3 ]
Traditional information retrieval techniques use a [[ histogram of keywords ]] as the << document representation >> but oral communication may offer additional indices such as the time and place of the rejoinder and the attendance .
USED-FOR
[ 6, 8, 11, 12 ]
An alternative index could be the << activity >> such as [[ discussing ]] , planning , informing , story-telling , etc. .
HYPONYM-OF
[ 9, 9, 6, 6 ]
An alternative index could be the activity such as [[ discussing ]] , << planning >> , informing , story-telling , etc. .
CONJUNCTION
[ 9, 9, 11, 11 ]
An alternative index could be the << activity >> such as discussing , [[ planning ]] , informing , story-telling , etc. .
HYPONYM-OF
[ 11, 11, 6, 6 ]
An alternative index could be the activity such as discussing , [[ planning ]] , << informing >> , story-telling , etc. .
CONJUNCTION
[ 11, 11, 13, 13 ]
An alternative index could be the << activity >> such as discussing , planning , [[ informing ]] , story-telling , etc. .
HYPONYM-OF
[ 13, 13, 6, 6 ]
An alternative index could be the activity such as discussing , planning , [[ informing ]] , << story-telling >> , etc. .
CONJUNCTION
[ 13, 13, 15, 15 ]
An alternative index could be the << activity >> such as discussing , planning , informing , [[ story-telling ]] , etc. .
HYPONYM-OF
[ 15, 15, 6, 6 ]
This paper addresses the problem of the << automatic detection >> of those [[ activities ]] in meeting situation and everyday rejoinders .
USED-FOR
[ 11, 11, 7, 8 ]
The format of the << corpus >> adopts the [[ Child Language Data Exchange System -LRB- CHILDES -RRB- ]] .
FEATURE-OF
[ 7, 14, 4, 4 ]
In this paper , we describe [[ data collection ]] , << transcription >> , word segmentation , and part-of-speech annotation of this corpus .
CONJUNCTION
[ 6, 7, 9, 9 ]
In this paper , we describe [[ data collection ]] , transcription , word segmentation , and part-of-speech annotation of this << corpus >> .
USED-FOR
[ 6, 7, 19, 19 ]
In this paper , we describe data collection , [[ transcription ]] , << word segmentation >> , and part-of-speech annotation of this corpus .
CONJUNCTION
[ 9, 9, 11, 12 ]
In this paper , we describe data collection , [[ transcription ]] , word segmentation , and part-of-speech annotation of this << corpus >> .
USED-FOR
[ 9, 9, 19, 19 ]
In this paper , we describe data collection , transcription , [[ word segmentation ]] , and << part-of-speech annotation >> of this corpus .
CONJUNCTION
[ 11, 12, 15, 16 ]
In this paper , we describe data collection , transcription , [[ word segmentation ]] , and part-of-speech annotation of this << corpus >> .
USED-FOR
[ 11, 12, 19, 19 ]
In this paper , we describe data collection , transcription , word segmentation , and [[ part-of-speech annotation ]] of this << corpus >> .
USED-FOR
[ 15, 16, 19, 19 ]
This paper shows how << dictionary word sense definitions >> can be analysed by applying a [[ hierarchy of phrasal patterns ]] .
USED-FOR
[ 14, 17, 4, 7 ]
An experimental << system >> embodying this [[ mechanism ]] has been implemented for processing definitions from the Longman Dictionary of Contemporary English .
PART-OF
[ 5, 5, 2, 2 ]
A property of this dictionary , exploited by the system , is that << it >> uses a [[ restricted vocabulary ]] in its word sense definitions .
USED-FOR
[ 16, 17, 13, 13 ]
A property of this dictionary , exploited by the system , is that it uses a [[ restricted vocabulary ]] in its << word sense definitions >> .
USED-FOR
[ 16, 17, 20, 22 ]
The structures generated by the experimental system are intended to be used for the << classification of new word senses >> in terms of the senses of words in the [[ restricted vocabulary ]] .
USED-FOR
[ 28, 29, 14, 18 ]
Thus the work reported addresses two [[ robustness problems ]] faced by current experimental << natural language processing systems >> : coping with an incomplete lexicon and with incomplete knowledge of phrasal constructions .
FEATURE-OF
[ 6, 7, 12, 15 ]
Thus the work reported addresses two << robustness problems >> faced by current experimental natural language processing systems : coping with an [[ incomplete lexicon ]] and with incomplete knowledge of phrasal constructions .
HYPONYM-OF
[ 20, 21, 6, 7 ]
Thus the work reported addresses two << robustness problems >> faced by current experimental natural language processing systems : coping with an incomplete lexicon and with [[ incomplete knowledge of phrasal constructions ]] .
HYPONYM-OF
[ 24, 28, 6, 7 ]
This paper presents a << word segmentation system >> in France Telecom R&D Beijing , which uses a unified [[ approach ]] to word breaking and OOV identification .
USED-FOR
[ 17, 17, 4, 6 ]
This paper presents a word segmentation system in France Telecom R&D Beijing , which uses a unified [[ approach ]] to << word breaking >> and OOV identification .
USED-FOR
[ 17, 17, 19, 20 ]
This paper presents a word segmentation system in France Telecom R&D Beijing , which uses a unified [[ approach ]] to word breaking and << OOV identification >> .
USED-FOR
[ 17, 17, 22, 23 ]
This paper presents a word segmentation system in France Telecom R&D Beijing , which uses a unified approach to [[ word breaking ]] and << OOV identification >> .
CONJUNCTION
[ 19, 20, 22, 23 ]
The system participated in all the tracks of the << segmentation bakeoff >> -- [[ PK-open ]] , PK-closed , AS-open , AS-closed , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
HYPONYM-OF
[ 12, 12, 9, 10 ]
The system participated in all the tracks of the segmentation bakeoff -- [[ PK-open ]] , << PK-closed >> , AS-open , AS-closed , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
CONJUNCTION
[ 12, 12, 14, 14 ]
The system participated in all the tracks of the << segmentation bakeoff >> -- PK-open , [[ PK-closed ]] , AS-open , AS-closed , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
HYPONYM-OF
[ 14, 14, 9, 10 ]
The system participated in all the tracks of the segmentation bakeoff -- PK-open , [[ PK-closed ]] , << AS-open >> , AS-closed , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
CONJUNCTION
[ 14, 14, 16, 16 ]
The system participated in all the tracks of the << segmentation bakeoff >> -- PK-open , PK-closed , [[ AS-open ]] , AS-closed , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
HYPONYM-OF
[ 16, 16, 9, 10 ]
The system participated in all the tracks of the segmentation bakeoff -- PK-open , PK-closed , [[ AS-open ]] , << AS-closed >> , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
CONJUNCTION
[ 16, 16, 18, 18 ]
The system participated in all the tracks of the << segmentation bakeoff >> -- PK-open , PK-closed , AS-open , [[ AS-closed ]] , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
HYPONYM-OF
[ 18, 18, 9, 10 ]
The system participated in all the tracks of the segmentation bakeoff -- PK-open , PK-closed , AS-open , [[ AS-closed ]] , << HK-open >> , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
CONJUNCTION
[ 18, 18, 20, 20 ]
The system participated in all the tracks of the << segmentation bakeoff >> -- PK-open , PK-closed , AS-open , AS-closed , [[ HK-open ]] , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
HYPONYM-OF
[ 20, 20, 9, 10 ]
The system participated in all the tracks of the segmentation bakeoff -- PK-open , PK-closed , AS-open , AS-closed , [[ HK-open ]] , << HK-closed >> , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
CONJUNCTION
[ 20, 20, 22, 22 ]
The system participated in all the tracks of the << segmentation bakeoff >> -- PK-open , PK-closed , AS-open , AS-closed , HK-open , [[ HK-closed ]] , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
HYPONYM-OF
[ 22, 22, 9, 10 ]
The system participated in all the tracks of the segmentation bakeoff -- PK-open , PK-closed , AS-open , AS-closed , HK-open , [[ HK-closed ]] , << MSR-open >> and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
CONJUNCTION
[ 22, 22, 24, 24 ]
The system participated in all the tracks of the << segmentation bakeoff >> -- PK-open , PK-closed , AS-open , AS-closed , HK-open , HK-closed , [[ MSR-open ]] and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
HYPONYM-OF
[ 24, 24, 9, 10 ]
The system participated in all the tracks of the segmentation bakeoff -- PK-open , PK-closed , AS-open , AS-closed , HK-open , HK-closed , [[ MSR-open ]] and << MSR - closed >> -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
CONJUNCTION
[ 24, 24, 26, 28 ]
The system participated in all the tracks of the << segmentation bakeoff >> -- PK-open , PK-closed , AS-open , AS-closed , HK-open , HK-closed , MSR-open and [[ MSR - closed ]] -- and achieved the state-of-the-art performance in MSR-open , MSR-close and PK-open tracks .
HYPONYM-OF
[ 26, 28, 9, 10 ]
The system participated in all the tracks of the segmentation bakeoff -- PK-open , PK-closed , AS-open , AS-closed , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in [[ MSR-open ]] , << MSR-close >> and PK-open tracks .
CONJUNCTION
[ 36, 36, 38, 38 ]
The system participated in all the tracks of the segmentation bakeoff -- PK-open , PK-closed , AS-open , AS-closed , HK-open , HK-closed , MSR-open and MSR - closed -- and achieved the state-of-the-art performance in MSR-open , [[ MSR-close ]] and << PK-open >> tracks .
CONJUNCTION
[ 38, 38, 40, 40 ]
This paper describes a [[ method ]] of << interactively visualizing and directing the process of translating >> a sentence .
USED-FOR
[ 4, 4, 6, 13 ]
The [[ method ]] allows a user to explore a << model >> of syntax-based statistical machine translation -LRB- MT -RRB- , to understand the model 's strengths and weaknesses , and to compare it to other MT systems .
USED-FOR
[ 1, 1, 8, 8 ]
The method allows a user to explore a [[ model ]] of << syntax-based statistical machine translation -LRB- MT -RRB- >> , to understand the model 's strengths and weaknesses , and to compare it to other MT systems .
USED-FOR
[ 8, 8, 10, 16 ]
The method allows a user to explore a model of syntax-based statistical machine translation -LRB- MT -RRB- , to understand the model 's strengths and weaknesses , and to compare [[ it ]] to other << MT systems >> .
COMPARE
[ 30, 30, 33, 34 ]
Using this [[ visualization method ]] , we can find and address conceptual and practical problems in an << MT system >> .
USED-FOR
[ 2, 3, 16, 17 ]
A [[ method ]] of << sense resolution >> is proposed that is based on WordNet , an on-line lexical database that incorporates semantic relations -LRB- synonymy , antonymy , hyponymy , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
USED-FOR
[ 1, 1, 3, 4 ]
A << method >> of sense resolution is proposed that is based on [[ WordNet ]] , an on-line lexical database that incorporates semantic relations -LRB- synonymy , antonymy , hyponymy , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
USED-FOR
[ 11, 11, 1, 1 ]
A method of sense resolution is proposed that is based on [[ WordNet ]] , an << on-line lexical database >> that incorporates semantic relations -LRB- synonymy , antonymy , hyponymy , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
HYPONYM-OF
[ 11, 11, 14, 16 ]
A method of sense resolution is proposed that is based on << WordNet >> , an on-line lexical database that incorporates [[ semantic relations ]] -LRB- synonymy , antonymy , hyponymy , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
PART-OF
[ 19, 20, 11, 11 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates << semantic relations >> -LRB- [[ synonymy ]] , antonymy , hyponymy , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
HYPONYM-OF
[ 22, 22, 19, 20 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates semantic relations -LRB- [[ synonymy ]] , << antonymy >> , hyponymy , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
CONJUNCTION
[ 22, 22, 24, 24 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates << semantic relations >> -LRB- synonymy , [[ antonymy ]] , hyponymy , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
HYPONYM-OF
[ 24, 24, 19, 20 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates semantic relations -LRB- synonymy , [[ antonymy ]] , << hyponymy >> , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
CONJUNCTION
[ 24, 24, 26, 26 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates << semantic relations >> -LRB- synonymy , antonymy , [[ hyponymy ]] , meronymy , causal and troponymic entailment -RRB- as labeled pointers between word senses .
HYPONYM-OF
[ 26, 26, 19, 20 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates semantic relations -LRB- synonymy , antonymy , [[ hyponymy ]] , << meronymy >> , causal and troponymic entailment -RRB- as labeled pointers between word senses .
CONJUNCTION
[ 26, 26, 28, 28 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates << semantic relations >> -LRB- synonymy , antonymy , hyponymy , [[ meronymy ]] , causal and troponymic entailment -RRB- as labeled pointers between word senses .
HYPONYM-OF
[ 28, 28, 19, 20 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates semantic relations -LRB- synonymy , antonymy , hyponymy , [[ meronymy ]] , << causal and troponymic entailment >> -RRB- as labeled pointers between word senses .
CONJUNCTION
[ 28, 28, 30, 33 ]
A method of sense resolution is proposed that is based on WordNet , an on-line lexical database that incorporates << semantic relations >> -LRB- synonymy , antonymy , hyponymy , meronymy , [[ causal and troponymic entailment ]] -RRB- as labeled pointers between word senses .
HYPONYM-OF
[ 30, 33, 19, 20 ]
With [[ WordNet ]] , it is easy to retrieve sets of << semantically related words >> , a facility that will be used for sense resolution during text processing , as follows .
USED-FOR
[ 1, 1, 10, 12 ]
With WordNet , it is easy to retrieve sets of [[ semantically related words ]] , a facility that will be used for << sense resolution >> during text processing , as follows .
USED-FOR
[ 10, 12, 21, 22 ]