Recent Named Entity Recognition and Classification techniques: A systematic review

The rapid increase in the flow rate of published digital information in all disciplines has resulted in a pressing need for techniques that can simplify the use of this information. The chemistry literature is very rich with information about chemical entities. Extracting molecules and their related properties and activities from the scientific literature to “text mine” these extracted data and determine contextual relationships helps research scientists, particularly those in drug development. One of the most important challenges in chemical text mining is the recognition of chemical entities mentioned in the texts. In this review, the authors briefly introduce the fundamental concepts of chemical literature mining, the textual contents of chemical documents, and the methods of naming chemicals in documents. We sketch out dictionary-based, rule-based and machine learning, as well as hybrid chemical named entity recognition approaches with their applied solutions. We end with an outlook on the pros and cons of these approaches and the types of chemical entities extracted.

/pdf/chemical-named-entities-recognition-a-review-on-approaches-9utpuz3uj7.pdf

Chemical named entities recognition: a review on approaches and applications

Even when the role of a conversational agent is well known users persist in confronting them with Out-of-Domain input. This often results in inappropriate feedback, leaving the user unsatisfied. In this paper we explore the automatic creation/enrichment of conversational agents’ knowledge bases by taking advantage of natural language interactions present in the Web, such as movies subtitles. Thus, we introduce Filipe, a chatbot that answers users’ request by taking advantage of a corpus of turns obtained from movies subtitles (the Subtle corpus). Filipe is based on Say Something Smart, a tool responsible for indexing a corpus of turns and selecting the most appropriate answer, which we fully describe in this paper. Moreover, we show how this corpus of turns can help an existing conversational agent to answer Out-of-Domain interactions. A preliminary evaluation is also presented.

http://www.inesc-id.pt/pt/indicadores/Ficheiros/10327.pdf

Luke, I am Your Father: Dealing with Out-of-Domain Requests by Using Movies Subtitles

Exploring the chemical and biological space covered by patent applications is crucial in early-stage medicinal chemistry activities. Patent analysis can provide understanding of compound prior art, novelty checking, validation of biological assays, and identification of new starting points for chemical exploration. Extracting chemical and biological entities from patents through manual extraction by expert curators can take substantial amount of time and resources. Text mining methods can help to ease this process. To validate the performance of such methods, a manually annotated patent corpus is essential. In this study we have produced a large gold standard chemical patent corpus. We developed annotation guidelines and selected 200 full patents from the World Intellectual Property Organization, United States Patent and Trademark Office, and European Patent Office. The patents were pre-annotated automatically and made available to four independent annotator groups each consisting of two to ten annotators. The annotators marked chemicals in different subclasses, diseases, targets, and modes of action. Spelling mistakes and spurious line break due to optical character recognition errors were also annotated. A subset of 47 patents was annotated by at least three annotator groups, from which harmonized annotations and inter-annotator agreement scores were derived. One group annotated the full set. The patent corpus includes 400,125 annotations for the full set and 36,537 annotations for the harmonized set. All patents and annotated entities are publicly available at www.biosemantics.org.

/pdf/annotated-chemical-patent-corpus-a-gold-standard-for-text-35i8jf1wb4.pdf

Annotated Chemical Patent Corpus: A Gold Standard for Text Mining

https://www.site.uottawa.ca/~diana/publications/diman_journal_paper.pdf

Prior and contextual emotion of words in sentential context

This paper presents a pioneering work on building a Named Entity Recognition system for the Mongolian language, with an agglutinative morphology and a subject-object-verb word order. Our work explores the fittest feature set from a wide range of features and a method that refines machine learning approach using gazetteers with approximate string matching, in an effort for robust handling of out-of-vocabulary words. As well as we tried to apply various existing machine learning methods and find optimal ensemble of classifiers based on genetic algorithm. The classifiers uses different feature representations. The resulting system constitutes the first-ever usable software package for Mongolian NER, while our experimental evaluation will also serve as a much-needed basis of comparison for further research.

/pdf/named-entity-recognition-for-mongolian-language-fvb4ez48ua.pdf

Named Entity Recognition for Mongolian Language

This paper presents a trigram part-of-speech tagger for Mongolian which is important to a large corpus because hand annotating is complicated task and assigning POS tags to each word is very time consuming.

Mongolian Trigram Part-of-Speech Tagger

This paper describes the development of finite state morphological transducer for Mongolian and presents some issues in Mongolian morphology, linguistic issues encountered and how they were dealt with. The work done here includes all the morphophonological rules needed for all Mongolian nominal and verb. Nominal morphotactic is implemented completely and verbal morphotactic covers one level continuation lexica. An evaluation is done via analysis on two separate corpora, which shows high-level and medium-level coverage respectively. It is more elaborate and accurate than previous implementations of its kinds.

Morphological Transducer for Mongolian

Goal of research work is to determine the most important features on predicting the grade of the General Entrance Exam (GEE). The features are high school student’s grade, personal information, and state exam results. We collected 96,827 high school students data and compared the F1 measures with different classification techniques such as decision tree, logistic regression, artificial neural network, and support vector machine. Among these techniques, the SVM provided the best F1 measure that is 0.70.

Predicting Grade of General Entrance Exam Using Machine Learning Techniques

As information flow in the electronic online environment is increasing, the need for automated processing is increasing. Every minute, 456,000 tweets were written on Twitter. The data in the Mongolian language is already joined in that stream and created their own space since a long time ago. However, in the Mongolian language, there is a lack of research, language resources, and an implemented system by now. In this work, we tried to classify the document level sentiment polarity of Mongolian tweets based on the RNN method of deep learning. The model was tested at the 62,917 SemEval English data, the highest F1 score of 64.7.

Zoljargal Munkhjargal

Papers

Named Entity Recognition for Mongolian Language

Mongolian Trigram Part-of-Speech Tagger

Morphological Transducer for Mongolian

Predicting Grade of General Entrance Exam Using Machine Learning Techniques

Sentiment Analysis for Mongolian Tweets with RNN