Summary of questionnaire.
| Input Contexts (IC) . | Citing Intent . | Candidates No. . | Topic of Candidate . | Analyzer’s Decision (AD) . | Relevancy . | Input Contexts No. . | Citing Intent . | Candidates (CAN) . | Topic of Candidate . | Decisions . | Relevancy . |
|---|---|---|---|---|---|---|---|---|---|---|---|
| IC1 | Techniques about sentence alignment | CAN1 | Text analysis | AD1: No | Not Relevant | IC6 | Facial modeling and drawbacks | CAN1 | Image registration | AD1: No | Not Relevant |
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN2 | Machine translation or parameters estimation | AD1: Yes | Weakly Relevant | CAN2 | Facial modeling | AD1: Yes | Strongly Relevant | ||||
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN3 | English-Chinese alignment | AD1: No | Weakly Relevant | CAN3 | Hierarchical motion estimation | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| CAN4 | Word correspondence algorithm | AD1: No | Weakly Relevant | CAN4 | Optical flow constraint | AD1: Yes | Weakly Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| CAN5 | Noun phrase alignment | AD1: No | Weakly Relevant | CAN5 | Facial model | AD1: No | Not Relevant | ||||
| AD2: Yes | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| IC2 | Noun phrase parsing | CAN1 | Part-of-speech tagger | AD1: Yes | Weakly Relevant | IC7 | Limitation of FACS approach | CAN1 | Facial modeling | AD1: No | Weakly Relevant |
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN2 | Rule-based parser | AD1: No | Not Relevant | CAN2 | Facial modeling and limitation of FACS | AD1: Yes | Strongly Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: Yes | ||||||||||
| CAN3 | Anaphora resolution | AD1: No | Not Relevant | CAN3 | Facial modeling | AD1: Yes | Weakly Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN4 | Formalism for parsing grammar statements | AD1: No | Not Relevant | CAN4 | Analysis of facial models | AD1: No | Weakly Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: Yes | ||||||||||
| CAN5 | Analysis of word association norm | AD1: No | Weakly Relevant | CAN5 | Image motion | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| IC3 | Part-of-speech tagger | CAN1 | Part-of-speech tagger | AD1: Yes | Strongly Relevant | IC8 | Maximum likelihoodlinear regression (MLLR) | CAN1 | MLLR | AD1: Yes | Strongly Relevant |
| AD2: Yes | AD2: Yes | ||||||||||
| AD3: Yes | AD3: Yes | ||||||||||
| CAN2 | Noun phrase tagger | AD1: No | Weakly Relevant | CAN2 | Maximum aposteriori estimation | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| CAN3 | Rule-based parser | AD1: No | Not Relevant | CAN3 | Hidden Markov model | AD1: No | Weakly Relevant | ||||
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN4 | Rule-based extraction of linguistic knowledge | AD1: No | Not Relevant | CAN4 | New covariance matrix | AD1: No | Weakly Relevant | ||||
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN5 | Case study of part-of-speech taggers | AD1: No | Not Relevant | CAN5 | Speech recognition | AD1: No | Weakly Relevant | ||||
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| IC4 | Sentence parser | CAN1 | Theoretical and empirical study on tree representation | AD1: No | Not Relevant | IC9 | Vector quantization | CAN1 | Latent dirichlet allocation (LDA) | AD1: No | Not Relevant |
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN2 | Text-chunking | AD1: No | Weakly Relevant | CAN2 | Matrix factorization | AD1: No | Weakly Relevant | ||||
| AD2: Yes | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN3 | Bilingual alignment | AD1: No | Not Relevant | CAN3 | Probabilistic latent semantic analysis (PLSA) | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN4 | tatistical parser | AD1: No | Weakly Relevant | CAN4 | PLSA | AD1: No | Not Relevant | ||||
| AD2: Yes | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN5 | Machine translation | AD1: No | Not Relevant | CAN5 | Latent variable models | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| IC5 | Bilingual alignment | CAN1 | Word-sense disambiguation | AD1: No | Not Relevant | IC10 | Non-negative matrix factorization (NMF) | CAN1 | NMF | AD1: No | Strongly Relevant |
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: Yes | ||||||||||
| CAN2 | Word-sense disambiguation | AD1: No | Not Relevant | CAN2 | LDA | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN3 | Word-sense disambiguation | AD1: No | Not Relevant | CAN3 | PLSA | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN4 | Bilingual word coding | AD1: Yes | Strongly Relevant | CAN4 | Auto-encoder with new training technique | AD1: No | Not Relevant | ||||
| AD2: Yes | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| CAN5 | Bilingual alignment | AD1: Yes | Strongly Relevant | CAN5 | Matrix decomposition on an over-complete basis | AD1: No | Not Relevant | ||||
| AD2: Yes | AD2: No | ||||||||||
| AD3: No | AD3: No |
| Input Contexts (IC) . | Citing Intent . | Candidates No. . | Topic of Candidate . | Analyzer’s Decision (AD) . | Relevancy . | Input Contexts No. . | Citing Intent . | Candidates (CAN) . | Topic of Candidate . | Decisions . | Relevancy . |
|---|---|---|---|---|---|---|---|---|---|---|---|
| IC1 | Techniques about sentence alignment | CAN1 | Text analysis | AD1: No | Not Relevant | IC6 | Facial modeling and drawbacks | CAN1 | Image registration | AD1: No | Not Relevant |
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN2 | Machine translation or parameters estimation | AD1: Yes | Weakly Relevant | CAN2 | Facial modeling | AD1: Yes | Strongly Relevant | ||||
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN3 | English-Chinese alignment | AD1: No | Weakly Relevant | CAN3 | Hierarchical motion estimation | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| CAN4 | Word correspondence algorithm | AD1: No | Weakly Relevant | CAN4 | Optical flow constraint | AD1: Yes | Weakly Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| CAN5 | Noun phrase alignment | AD1: No | Weakly Relevant | CAN5 | Facial model | AD1: No | Not Relevant | ||||
| AD2: Yes | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| IC2 | Noun phrase parsing | CAN1 | Part-of-speech tagger | AD1: Yes | Weakly Relevant | IC7 | Limitation of FACS approach | CAN1 | Facial modeling | AD1: No | Weakly Relevant |
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN2 | Rule-based parser | AD1: No | Not Relevant | CAN2 | Facial modeling and limitation of FACS | AD1: Yes | Strongly Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: Yes | ||||||||||
| CAN3 | Anaphora resolution | AD1: No | Not Relevant | CAN3 | Facial modeling | AD1: Yes | Weakly Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN4 | Formalism for parsing grammar statements | AD1: No | Not Relevant | CAN4 | Analysis of facial models | AD1: No | Weakly Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: Yes | ||||||||||
| CAN5 | Analysis of word association norm | AD1: No | Weakly Relevant | CAN5 | Image motion | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| IC3 | Part-of-speech tagger | CAN1 | Part-of-speech tagger | AD1: Yes | Strongly Relevant | IC8 | Maximum likelihoodlinear regression (MLLR) | CAN1 | MLLR | AD1: Yes | Strongly Relevant |
| AD2: Yes | AD2: Yes | ||||||||||
| AD3: Yes | AD3: Yes | ||||||||||
| CAN2 | Noun phrase tagger | AD1: No | Weakly Relevant | CAN2 | Maximum aposteriori estimation | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| CAN3 | Rule-based parser | AD1: No | Not Relevant | CAN3 | Hidden Markov model | AD1: No | Weakly Relevant | ||||
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN4 | Rule-based extraction of linguistic knowledge | AD1: No | Not Relevant | CAN4 | New covariance matrix | AD1: No | Weakly Relevant | ||||
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN5 | Case study of part-of-speech taggers | AD1: No | Not Relevant | CAN5 | Speech recognition | AD1: No | Weakly Relevant | ||||
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| IC4 | Sentence parser | CAN1 | Theoretical and empirical study on tree representation | AD1: No | Not Relevant | IC9 | Vector quantization | CAN1 | Latent dirichlet allocation (LDA) | AD1: No | Not Relevant |
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN2 | Text-chunking | AD1: No | Weakly Relevant | CAN2 | Matrix factorization | AD1: No | Weakly Relevant | ||||
| AD2: Yes | AD2: Yes | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN3 | Bilingual alignment | AD1: No | Not Relevant | CAN3 | Probabilistic latent semantic analysis (PLSA) | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN4 | tatistical parser | AD1: No | Weakly Relevant | CAN4 | PLSA | AD1: No | Not Relevant | ||||
| AD2: Yes | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN5 | Machine translation | AD1: No | Not Relevant | CAN5 | Latent variable models | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| IC5 | Bilingual alignment | CAN1 | Word-sense disambiguation | AD1: No | Not Relevant | IC10 | Non-negative matrix factorization (NMF) | CAN1 | NMF | AD1: No | Strongly Relevant |
| AD2: No | AD2: Yes | ||||||||||
| AD3: No | AD3: Yes | ||||||||||
| CAN2 | Word-sense disambiguation | AD1: No | Not Relevant | CAN2 | LDA | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN3 | Word-sense disambiguation | AD1: No | Not Relevant | CAN3 | PLSA | AD1: No | Not Relevant | ||||
| AD2: No | AD2: No | ||||||||||
| AD3: No | AD3: No | ||||||||||
| CAN4 | Bilingual word coding | AD1: Yes | Strongly Relevant | CAN4 | Auto-encoder with new training technique | AD1: No | Not Relevant | ||||
| AD2: Yes | AD2: No | ||||||||||
| AD3: Yes | AD3: No | ||||||||||
| CAN5 | Bilingual alignment | AD1: Yes | Strongly Relevant | CAN5 | Matrix decomposition on an over-complete basis | AD1: No | Not Relevant | ||||
| AD2: Yes | AD2: No | ||||||||||
| AD3: No | AD3: No |