Filters








135,124 Hits in 3.2 sec

Combining Logic and Aggregation for Answer Selection [chapter]

Ingo Glöckner
Lecture Notes in Computer Science  
MAVE (Multinet-based Answer Verification) is a system for answer validation which combines logic-based techniques and aggregation for identifying the correct answers in given sets of answer candidates.  ...  The paper explains the basic concepts underlying MAVE and also presents ablation studies which reveal the contribution of the proposed methods to the achieved quality of selection.  ...  It is not obvious how the logic-oriented techniques for answer validation and the redundancy-based techniques for selection must be combined in order to maximize selection quality.  ... 
doi:10.1007/978-3-540-85760-0_47 fatcat:ndymmwxrcbcephdx6yzpmknlli

RAVE: A Fast Logic-Based Answer Validator [chapter]

Ingo Glöckner
2009 Lecture Notes in Computer Science  
In this way parsing of the answers is avoided, yielding low validation/selection times. Machine learning is used for assigning local validation scores based on logical and shallow features.  ...  RAVE (Real-time Answer Validation Engine) is a logic-based answer validator/selector designed for real-time question answering.  ...  The aggregated support for k ∈ K q is then given by γ(k) = 1 − t∈T k µ(k, t) and for extracted answers by γ(a) = γ(κ(a)).  ... 
doi:10.1007/978-3-642-04447-2_56 fatcat:dxtftjy37zg2rkiuorg24dlglu

TAPAS: Weakly Supervised Table Parsing via Pre-training [article]

Jonathan Herzig, Paweł Krzysztof Nowak, Thomas Müller, Francesco Piccinno, Julian Martin Eisenschlos
2020 arXiv   pre-print
TAPAS trains from weak supervision, and predicts the denotation by selecting table cells and optionally applying a corresponding aggregation operator to such selection.  ...  In this paper, we present TAPAS, an approach to question answering over tables without generating logical forms.  ...  Given G c and the table values For cell selection examples, for which s is not populated, we train the model to select the cells in C. For scalar answer examples, table cells.  ... 
arXiv:2004.02349v2 fatcat:tbcfct2wgvcdvdeaqfu6iuiuoy

A Knowledge-based Approach for Answering Complex Questions in Persian [article]

Romina Etezadi, Mehrnoush Shamsfard
2021 arXiv   pre-print
Then Multilingual-BERT is used to select the logical form that best describes the input complex question syntactically and semantically.  ...  In this work, we handle multi-constraint and multi-hop questions by building their set of possible corresponding logical forms.  ...  Information required for each constraint is extracted from Farsbase (Sajadi et al., 2018) and is combined to create a set of possible logical forms.  ... 
arXiv:2107.02040v1 fatcat:qt7dhnxetrdw7dfcbvm3r7povu

Efficient Question Answering with Question Decomposition and Multiple Answer Streams [chapter]

Sven Hartrumpf, Ingo Glöckner, Johannes Leveling
2009 Lecture Notes in Computer Science  
IRSAW was introduced in 2007, by integrating the deep answer producer InSicht, several shallow answer producers, and a logical validator.  ...  The answer validator introduced in the previous year was replaced with the faster RAVE validator designed for logic-based answer validation under time constraints.  ...  Then a 'deep' answer score based on the combination of deep and shallow features is computed and the result is aggregated if it exceeds the quality threshold.  ... 
doi:10.1007/978-3-642-04447-2_49 fatcat:6grdchnn7va4vl5k7zoo2ncsly

Effective trust management through a hybrid logical and relational approach

Adam J. Lee, Ting Yu, Yann Le Gall
2010 Proceedings of the 5th ACM Symposium on Information, Computer and Communications Security - ASIACCS '10  
We describe a correct and complete procedure for compiling CTM credentials into dynamic views within a database, and show how the resulting system can be used to perform role membership checks or to enumerate  ...  We then propose a hybrid algorithm that leverages the logical ruleset and the underlying DBMS to efficiently enumerate the capabilities ascribed to a given user.  ...  The research described in this paper was supported by the National Science Foundation under awards CCF-0916015 and CCF-0914946.  ... 
doi:10.1145/1755688.1755710 dblp:conf/ccs/LeeYG10 fatcat:2aek5ng5drhwhk4e5pwjfb55te

Combining Logic and Machine Learning for Answering Questions [chapter]

Ingo Glöckner, Björn Pelzer
2009 Lecture Notes in Computer Science  
Its main innovation is the use of logic for simultaneously extracting answer bindings and validating the corresponding answers.  ...  LogAnswer is a logic-oriented question answering system developed by the AI research group at the University of Koblenz-Landau and by the IICS at the University of Hagen.  ...  Aggregation and Answer Selection The answer integration module computes a global score for each answer, based on the local score for each passage from which the answer was extracted [3] .  ... 
doi:10.1007/978-3-642-04447-2_47 fatcat:2huinbl3cbairivvjmg53pfity

Question Generation from SQL Queries Improves Neural Semantic Parsing

Daya Guo, Yibo Sun, Duyu Tang, Nan Duan, Jian Yin, Hong Chi, James Cao, Peng Chen, Ming Zhou
2018 Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing  
In addition, we observe that there is a logarithmic relationship between the accuracy of a semantic parser and the amount of training data.  ...  Thanks to the anonymous reviewers for their helpful comments and suggestions.  ...  Seq2SQL which extends Aug.PntNet by further learning two separate classifiers for SELECT aggregator and SELECT column.  ... 
doi:10.18653/v1/d18-1188 dblp:conf/emnlp/GuoSTDYCCCZ18 fatcat:i3kwgeokkvccdooqvg3hdkky6q

Question Generation from SQL Queries Improves Neural Semantic Parsing [article]

Daya Guo, Yibo Sun, Duyu Tang, Nan Duan, Jian Yin, Hong Chi, James Cao, Peng Chen, Ming Zhou
2018 arXiv   pre-print
In addition, we observe that there is a logarithmic relationship between the accuracy of a semantic parser and the amount of training data.  ...  Thanks to the anonymous reviewers for their helpful comments and suggestions.  ...  Seq2SQL which extends Aug.PntNet by further learning two separate classifiers for SELECT aggregator and SELECT column.  ... 
arXiv:1808.06304v2 fatcat:4jgswmw2pvfqbakowhxrdkedgq

Mathematical Tutorial System (MTS) Using Fuzzy Logic and Multi-Agent System

Gandhimathi K., Yogarani A., Sr.J.Arockia Jaya, Ponmathi M.
2017 International Journal of Innovative Research in Engineering & Management  
We use the combined approach of Fuzzy logic and Multiagent System (MAS).  ...  It also guides and monitors student progress. Our aim is to design an instructional planner for an intelligent tutoring system.  ...  So the Instructional Planner for an MTS must dynamically adapt to each students to teach the number and addition concepts. So we used the combined approach of fuzzy logic and multiagent systems.  ... 
doi:10.21276/ijirem.2017.4.2.1 fatcat:pcftswoa55dahinnjoozcmenxq

Enhancing Algebraic Query Relaxation with Semantic Similarity [article]

Lena Wiese
2012 arXiv   pre-print
Cooperative database systems support a database user by searching for answers that are closely related to his query and hence are informative answers.  ...  Moreover we propose some heuristics to assign a degree of similarity to each tuple of an answer table; this degree can help the user to determine whether this answer is relevant for him or not.  ...  An aggregation function can be used to condense the values in the answer table into one and then compare the values for the two answer tables.  ... 
arXiv:1210.6242v1 fatcat:pkgoonfow5cbvprzjqfyelfkwa

Universal approximation with uninorm-based fuzzy neural networks

Andre Lemos, Vladik Kreinovich, Walmir Caminhas, Fernando Gomide
2011 2011 Annual Meeting of the North American Fuzzy Information Processing Society  
We will describe the network topology and inference mechanism and show that the universal approximation property of this network is valid for a given choice of operators.  ...  In this paper we will show that the recently proposed fuzzy neural network based on weighted uninorms aggregations uniformly approximates any real functions on any compact set.  ...  Our answer: formulation.  ... 
doi:10.1109/nafips.2011.5752000 fatcat:co3rlqwq2nbg3hpryoav76fznq

Toward Agile BI By Using In-Memory Analytics

Mihaela MUNTEAN
2014 Informatică economică  
And, which is the best data model that can be used for enabling an agile BI solution?  ...  Also, the paper focuses on the main data models used by in-memory BI technologies and tries to answer following questions: Which are the main characteristics of an agile data model?  ...  But, we can make every data selection (any combination of year and region). For example, Figure 7b shows top 2 sellers for each city in 2012, only for Transilvania region and Muntenia region.  ... 
doi:10.12948/issn14531305/18.3.2014.03 fatcat:v7k5kwmedndojmqr32t4qdgarm

Thresholded Tabulation in a Fuzzy Logic Setting

Pascual Julián, Jesús Medina, Ginés Moreno, Manuel Ojeda-Aciego
2009 Electronical Notes in Theoretical Computer Science  
In this paper we propose an improved fuzzy query answering procedure for the so called multi-adjoint logic programming approach, which avoids the re-evaluation of goals and the generation of useless computations  ...  Fuzzy logic programming represents a flexible and powerful declarative paradigm amalgamating fuzzy logic and logic programming, for which there exists different promising approaches described in the literature  ...  values of L and conjunctions, disjunctions and aggregations.  ... 
doi:10.1016/j.entcs.2009.07.063 fatcat:tep6tbgsizbfplwpwfq2ujmyoq

Adaptive signature-based semantic selection of services with OWLS-MX3

Matthias Klusch, Patrick Kapahnke, Michael N. Huhns, Ryszard Kowlczyk, Zakaria Maamar, Rainer Unland
2012 Multiagent and Grid Systems  
It learns how to best combine logic-based, text similarity and ontological structure matching for hybrid semantic selection of OWL-S services to given queries.  ...  The matchmaker adapts its selection to changes in the semantic service landscape and ontologies by learning the respectively optimal weighted combination of different types of semantic similarities it  ...  The work reported in this paper has been supported by the German ministry for education and research (BMB+F) under project grant 01IW08005 (ISReal); this paper is an invited and updated version of [13  ... 
doi:10.3233/mgs-2012-0184 fatcat:xfuiij7axfhi7bf5tsi3nooglm
« Previous Showing results 1 — 15 out of 135,124 results