ChAnot: An Intelligent Annotation Tool for Indigenous and Highly Agglutinative Languages in Peru

ChAnot: An Intelligent Annotation Tool for Indigenous and Highly Agglutinative Languages in Peru Rodolfo Mercado-Gonzales, Jos´e Pereira-Noriega, Marco Sobrevilla, Arturo Oncevay Research Group on Artificial Intelligence (IA-PUCP) Departamento de Ingenier´ıa, Pontificia Universidad Cat´olica del Per´u, Lima, Peru {rmercado, jpereira, msobrevilla, arturo.oncevay}@pucp.edu.pe Abstract Linguistic corpus annotation is one of the most important phases for addressing Natural Language Processing (NLP) tasks, as these methods are deeply involved with corpus-based techniques. However, meta-data annotation is a highly laborious manual task. A supportive alternative requires the use of computational tools. They are likely to simplify some of these operations, while can be adjusted appropriately to the needs of particular language features at the same time. Therefore, this paper presents ChAnot, a web-based annotation tool developed for Peruvian indigenous and highly agglutinative languages, where Shipibo-Konibo was the case study. This new tool is able to support a diverse set of linguistic annotation tasks, such as morphological segmentation markup, POS-tag markup, among others. Also, it includes a suggestion engine based on historic and machine learning models, and a set of statistics about previous annotations. Keywords: Annotation Tool, Corpus Annotation, Peruvian Indigenous Languages, Shipibo-Konibo 1. Introduction The research field of Natural Language Processing (NLP) aims the analysis automation, representation and gen- eration of human language through computational tech- niques (Cambria and White, 2014). These methods usu- ally are based on supervised machine learning approaches, which employ the analysis of corpus, composed of anno- tated samples, and other strategies for reaching a particular NLP goal. The samples within a corpus need to be enriched with additional information or meta-data, plus the correct solution regarding the NLP task, generating an annotated linguistic corpus (M¨uller and Strube, 2006). Nevertheless, the development of an annotated linguistic corpus might be a highly time-consuming task. The pro- cess requires mainly human effort, which is performed by linguists or experts in a particular language. While the cor- pus is more specialized or the language is less widespread, the expertise requirements will increase. In that way, an easier specialized annotation tool is what is needed. (M¨uller and Strube, 2006). This paper introduces ChAnot, a new web-based annota- tion tool, which is focused in Peruvian indigenous and highly agglutinative languages. ChAnot enables the cor- pus annotation for various NLP tasks, such as Morpho- logical Analysis (lemma and morphological segmentation), Part-of-Speech tagging, Name Entity Recognition and Syn- tactic Analysis (using a BRAT1 interface (Stenetorp et al., 2012)). The interactive interface and architecture are ad- justed appropriately to the needs of the Peruvian indigenous languages and humans annotators. Moreover, ChAnot in- cludes a suggestion engine based in machine learning mod- els and a set of important statistics about historic annota- tion. The text below is organized as follow. Section 2 presents some existing annotation tools. Then, the main and dis- tinctive functionalities in ChAnot are detailed in Section 3. 1Available in: http://brat.nlplab.org/ Later, Section 4 presents briefly features regarding the Pe- ruvian indigenous languages, focusing in Shipibo-Konibo, which is the case study language. Finally, conclusions and future works …

Content truncated to 500 words