Semi-Automatic Development of KurdNet, The Kurdish WordNet
Article Structure
Abstract
Recently, we reported on our efforts to build the first prototype of KurdNet.
Introduction
WordNet (Fellbaum, 2010) has been used in numerous natural language processing tasks such as word sense disambiguation and information extraction with considerable success.
KurdNet: State-of-the-Art
In our previous work (Aliabadi et al., 2014), we described the steps that we have taken to build the first prototype of KurdNet.
KurdNet: Shortcomings
The current version of KurdNet is quite basic and therefore its applicability is very limited.
KurdNet: Extension Plan
4.1 Goals and Envisioned Outcomes
Summary
In this paper, we underlined the major shortcomings in the current KurdNet prototype and proposed a concrete plan to enrich the current prototype, so that it can be used in development of Kurdish language processing systems.
Topics
synsets
Appears in 23 sentences as: /SYNSET (1) SYNSET (1) Synset (4) synset (8) Synsets (2) synsets (12)
In Semi-Automatic Development of KurdNet, The Kurdish WordNet
- 0 Expand: in this model, the synsets are built in correspondence with the WordNet synsets and the semantic relations are directly imported.Page 2, “KurdNet: State-of-the-Art”
- 0 Merge: in this model, the synsets and relations are first built independently and then they are aligned with WordNet’s.Page 2, “KurdNet: State-of-the-Art”
- synsets ) that play a major role in the wordnets.Page 2, “KurdNet: State-of-the-Art”
- A sample synset is depicted in Figure 3.Page 2, “KurdNet: State-of-the-Art”
- <SYNSET> < | D>ENGZO-00008853-v</| D> <POS>v</POS> <SYNONYM> <LITERAL>shed<SENSE>4</SENSE></LITERAL> <LITERAL>moIt<SENSE>1</SENSE></LITERAL> <LITERAL>exuviate<SENSE>1</SENSE></LITERAL> <LITERAL>moult<SENSE>1</SENSE></LITERAL> <LITERAL>sIough<SENSE>1</SENSE></LITERAL> </SYNONYM> <|LR><TYPE>hypernym</TYPE>ENGZO-01471089-v</|LR> <|LR><TYPE>eng_derivative</TYPE>ENGZO-01245451-n</| LR> < | LR><TYP E>eng_derivative</TYP E>ENGZO-08844332-n</| LR> <|LR><TYPE>eng_derivative</TYPE>ENG20-12753095-n</| LR> <|LR><TYPE>eng_derivative</TYPE>ENGZO-12791455-n</| LR> <DEF>cast off hair, skin, horn, or feathers</DEF> <USAG E>out dog sheds every Spring</USAGE> <BCS>2</BCS> <DOMA|N>zooIogy</DOMAIN> <SUMO>Removing<TYPE>+</TYPE></SUMO> < /SYNSET >Page 2, “KurdNet: State-of-the-Art”
- Figure 3: A WordNet verb synset in XML (Vossen et al., 1998)Page 2, “KurdNet: State-of-the-Art”
- Since KurdNet follows the Expand model, it inherits most of Base Concepts’ structural properties, including: synsets and the lexical relations among them, POS, Domain, BCS, and SUMO.Page 2, “KurdNet: State-of-the-Art”
- Synset Alignments: for each synset in BC, its counterpart in KurdNet is defined semiautomatically.Page 2, “KurdNet: State-of-the-Art”
- Figure 4: An Illustration of a Synset in Base Concepts and its Maximal and Minimal Alignment Variants in KurdNet (Aliabadi et al., 2014)Page 3, “KurdNet: State-of-the-Art”
- Concepts (max) (min) Synset No .Page 3, “KurdNet: State-of-the-Art”
- Usage Examples: we have taken a corpus-assisted approach to speedup the process of providing usage examples for each aligned synset .Page 3, “KurdNet: State-of-the-Art”
See all papers in Proc. ACL 2014 that mention synsets.
See all papers in Proc. ACL that mention synsets.
Back to top.
WordNet
Appears in 18 sentences as: WordNet (9) wordnet (4) wordnets (4) WordNet’s (1)
In Semi-Automatic Development of KurdNet, The Kurdish WordNet
- WordNet (Fellbaum, 2010) has been used in numerous natural language processing tasks such as word sense disambiguation and information extraction with considerable success.Page 1, “Introduction”
- Kurdish is a less-resourced language for which, among other resources, no wordnet has been built yet.Page 1, “Introduction”
- 1. highlighted the main challenges in building a wordnet for the Kurdish language (including its inherent diversity and morphological complexity),Page 1, “KurdNet: State-of-the-Art”
- 2. built the first prototype of KurdNet, the Kurdish WordNet (see a summary below), andPage 1, “KurdNet: State-of-the-Art”
- There are two well-known models for building wordnets for a language (Vossen, 1998):Page 2, “KurdNet: State-of-the-Art”
- 0 Expand: in this model, the synsets are built in correspondence with the WordNet synsets and the semantic relations are directly imported.Page 2, “KurdNet: State-of-the-Art”
- 0 Merge: in this model, the synsets and relations are first built independently and then they are aligned with WordNet’s .Page 2, “KurdNet: State-of-the-Art”
- The expand model seems less complex and guarantees the highest degree of compatibility across different wordnets .Page 2, “KurdNet: State-of-the-Art”
- synsets) that play a major role in the wordnets .Page 2, “KurdNet: State-of-the-Art”
- Base Concepts (BC) is available on the Global WordNet Association (GWA)’s Web page2.Page 2, “KurdNet: State-of-the-Art”
- Figure 3: A WordNet verb synset in XML (Vossen et al., 1998)Page 2, “KurdNet: State-of-the-Art”
See all papers in Proc. ACL 2014 that mention WordNet.
See all papers in Proc. ACL that mention WordNet.
Back to top.
semantic relations
Appears in 6 sentences as: Semantic Relation (1) semantic relation (1) Semantic Relations (1) semantic relations (3)
In Semi-Automatic Development of KurdNet, The Kurdish WordNet
- 0 Expand: in this model, the synsets are built in correspondence with the WordNet synsets and the semantic relations are directly imported.Page 2, “KurdNet: State-of-the-Art”
- 3.3 Limited Support for Semantic Relation TypesPage 3, “KurdNet: Shortcomings”
- As shown in Table 2, there are several WordNet semantic relations for each syntactic categories.Page 3, “KurdNet: Shortcomings”
- The most important semantic relation in WordNet is Hyponymy and this relation is the only one support in KurdNet (Aliabadi et al., 2014).Page 3, “KurdNet: Shortcomings”
- o to winden the scope (i.e., including Kurmanji synsets), the coverage (i.e., going beyond Base Concepts) , and richness (supporting additional semantic relations ) of the current version.Page 4, “KurdNet: Extension Plan”
- Finding Semantic Relations Creating Graphical User InterfacePage 6, “Summary”
See all papers in Proc. ACL 2014 that mention semantic relations.
See all papers in Proc. ACL that mention semantic relations.
Back to top.
