Sentence Boundary Detection is a pre-processing step for any Natural Language Processing application. Various algorithms have been used to achieve Sentence Boundary Detection or Disambiguation in different languages. In this paper, a rule based method is proposed and tested to achieve Sentence Boundary Detection for Kannada Language. Kannada being a grammatically rich Indian language is analyzed based on semantics and tested with a 227K bytes corpus. The code is written in C using wide characters, with support for Unicode. Results showed 99.2% success in detecting sentence boundary.
In this era of digitalization, everything is smart and digitalized. All the documents are presented, prepared and shared as soft copies. Classifying those soft copy documents is gaining an important insight in recent times. It is attracting digital word with its impact in different fields like spam filtering, email routing, language identification, genre classification, sentimental analysis, readability assessment. Classifying documents that are available online using smart techniques helps different business. The easiest and efficient way of doing it is through machine learning and it makes human work much easier. To perform classification of document more statistically, documents should be given in a much understandable format to the machine learning classifier. In this report, I'm discussing the types of feature depending on which an document can be classified and later represented. Record arrangement or classifying the documents is the purpose of document collection and classifications based upon the information it consists off and features that it contains. Record arrangement is a huge learning issue that is at the center of numerous data executives and recovery. Document grouping plays an important role in different applications that help with sorting out, ordering, looking and briefly speaking to a lot of data. In this report, we will be discussing the uses of document classification and important steps used for classifying the document or text by considering a small use case to know how document classification is done, basic steps of document classification, processing and analyzing the documents that are collected. We have considered two different categories of data sets for classification and analysis. The problem statement here is to distinguish those two documents where one is Rhyme document and each rhyme is taken as a single file and the other is normal sentences that are a Non-Rhyme document that contains normal Wikipedia text where few statements of Wikipedia is considered as a single file. The precise objective of my project is to develop scalable and efficient document classification project that classifies the document more precisely depending on the feature that it contains and to know the basic techniques that are used for the document a classification like, data collection, data cleaning, pre-processing and constructing an ML model and applying the ML algorithm. Another objective of the project is to work on machine learning concepts and to get insight into different classification algorithms with the help of this case study.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.