A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2015; you can also visit the original URL.
The file type is application/pdf.
Filters
Maxprob and categorization of queries based on linguistic features
2007
Proceedings of the ACM first Ph.D. workshop in CIKM on - PIKM '07
Preserved Fulltext
Our fusing technique is based on queries that are classified using some automatically extracted linguistic features [11] . ...
The rank list of relevant documents provided by the systems are divided into segments and used in a training/testing process to detect the systems to fuse. ...
The selection is based on clusters of queries that take into account variability and homogeneity in query features. ...
doi:10.1145/1316874.1316885
dblp:conf/cikm/KompaoreM07
fatcat:puzats57hrhzpmwxkgoznjr2wy
TFW, DamnGina, Juvie, and Hotsie-Totsie: On the Linguistic and Social Aspects of Internet Slang
[article]
2017
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In this work, we use UrbanDictionary to conduct the first large-scale linguistic analysis of slang and its social aspects on the Internet to yield insights into this variety of language that is increasingly ...
Analyzing tens of thousands of such slang words reveals that the majority of slang on the Internet belongs to two major categories: sex and drugs. ...
Furthermore the model using character-ngram features significantly outperforms the morpheme based model. ...
arXiv:1712.08291v1
fatcat:d5atsrawsrczpdgtmgwice5esu
Type- and Content-Driven Synthesis of SQL Queries from Natural Language
[article]
2017
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
We have implemented the proposed technique in a tool called Sqlizer and evaluate it on three different databases. ...
Our experiments show that the desired query is ranked within the top 5 candidates in close to 90% of the cases. ...
well across multiple databases and outperforms a state-of-the-art NLIDB system. ...
arXiv:1702.01168v1
fatcat:r6i3h7f3enaz7javsf7754kblm
SQLizer: query synthesis from natural language
2017
Proceedings of the ACM on Programming Languages
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We evaluate our approach on over 450 natural language queries concerning three different databases, namely MAS, IMDB, and YELP. ...
At the core of our technique is a new NL-based program synthesis methodology that combines semantic parsing techniques from the NLP community with type-directed program synthesis and automated program ...
This material is based on research sponsored by DARPA under agreement number #8750-14-2-0270. The U.S. ...
doi:10.1145/3133887
dblp:journals/pacmpl/Yaghmazadeh0DD17
fatcat:vlqlptvygbbijhvcibq5j5tlk4
Will this Question be Answered? Question Filtering via Answer Model Distillation for Efficient Question Answering
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Our experiments on three popular QA datasets and one industrial QA benchmark demonstrate the ability of our question models to approximate the Precision/Recall curves of the target QA system well. ...
This is based on an interesting new finding: the answer confidence scores of state-of-the-art QA systems can be approximated well by models solely using the input question text. ...
Acknowledgements We thank the anonymous reviewers and metareviewer for their valuable suggestions. We thank Thuy Vu for developing and sharing the human annotated data used in the AQAD dataset. ...
arXiv:2109.07009v1
fatcat:a2tg34evkrdnjjjdakvqng2q3i
Automated synthesis of data extraction and transformation programs
[article]
2017
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
This score is calculated based on a set of pre-defined features. ...
To assess how Sqlizer performs on different classes of queries, we manually categorize the benchmarks into four groups based on the characteristics of their corresponding SQL query. ...
For example, we experiments on 455 queries from three different databases shows that Sqlizer ranks the desired query as top one in 78% of the cases and among top 5 in ∼ 90% of the time. ...
doi:10.15781/t2zk56545
fatcat:guknbbfyfnacnmgfxvs2gqcamu
Will this Question be Answered? Question Filtering via Answer Model Distillation for Efficient Question Answering
2021
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
unpublished
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Our experiments on three popular QA datasets and one industrial QA benchmark demonstrate the ability of our question models to approximate the Precision/Recall curves of the target QA system well. ...
This is based on an interesting new finding: the answer confidence scores of state-of-the-art QA systems can be approximated well by models solely using the input question text. ...
Acknowledgements We thank the anonymous reviewers and metareviewer for their valuable suggestions. We thank Thuy Vu for developing and sharing the human annotated data used in the AQAD dataset. ...
doi:10.18653/v1/2021.emnlp-main.583
fatcat:ltajh5krtbfybjnntyerlvqpsy