Named Entity Recognition for the Indonesian Language: Combining Contextual, Morphological and Part-of-Speech Features into a Knowledge Engineering Approach

Budi, Indra; Bressan, Stéphane; Wahyudi, Gatot; Hasibuan, Zainal A.; Nazief, Bobby Achirul Awal

doi:10.1007/11563983_7

Cited by 28 publications

(22 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Sebagai contoh, aplikasi Arabic Named Entity Recognition (NERA) dibina khusus bagi domain jenayah dalam bahasa Arab (Asharef, 2012). Selain itu, penyelidikan NER turut giat dilakukan ke atas bahasa Indonesia, Melayu dan Iban Budi, Bressan & Wahyudi, 2005;Fong, Ranaivo-Malançon & Wee, 2011, Shahrul Azman et al, 2018. Namun, buat masa ini masih belum ada lagi penyelidikan yang dibina khusus bagi mengenalpasti jenis NER dalam domain jenayah bahasa Melayu.…”

Section: Latar Belakang Kajianunclassified

Pendekatan Teknik Pengecaman Entiti Nama Bagi Capaian Berita Jenayah Bahasa Melayu (Named Entity Recognition Approach for Malay Crime News Retrieval)

Saad

Mansor

2018

GEMA

View full text Add to dashboard Cite

Pengekstrakan maklumat merupakan satu proses bagi mendapatkan konsep penting dalam mewakili kandungan teks dari dokumen yang tidak berstruktur. Pada masa kini, terdapat banyak dokumen yang tidak berstruktur seperti teks berita, artikel blog, forum, tweet serta mikro blog dari rangkaian sosial. Dokumen-dokumen ini amat sukar untuk difahami oleh komputer. Oleh itu, kajian berkaitan pengekstrakan maklumat menjadi sangat penting bagi mengatasi permasalah ini. Salah satu teknik pengekstrakan yang banyak digunakan ialah pengecaman entiti nama. Kajian ini dijalankan bagi mengimplementasikan teknik pengecaman entiti nama dari sumber dokumen berita jenayah bahasa Melayu. Objektif utama kajian ini adalah untuk membangunkan sistem prototaip model pengekstrakan maklumat berita jenayah dalam bahasa Melayu dengan menggunakan teknik pengecaman entiti nama melalui pendekatan berasaskan peraturan. Kajian ini dilakukan dengan mewujudkan korpus berita jenayah dalam bahasa Melayu yang diperolehi dari sumber arkib berita BERNAMA. Korpus ini kemudiannya diteliti secara manual oleh pakar bahasa bagi mengecam entiti nama seperti individu, organisasi, lokasi, tarikh, masa, kewangan, peratusan, jenayah dan senjata. Dalam masa yang sama, sistem prototaip dibangunkan serta diuji dengan korpus yang sama dan hasil dari pengujian ini dibandingkan dengan keputusan pakar. Secara keseluruhannya, ujian sistem prototaip ini menunjukkan hasil yang baik dengan nilai dapatan bagi recall sebanyak 78.67%, manakala bagi precision ialah sebanyak 71.11% dan F-measure sebanyak 74.7%. Hasil dari kajian ini diharap dapat menyumbang kepada pengetahuan mengenai keberkesanan teknik pengecaman entiti nama bagi berita jenayah bahasa Melayu dan seterusnya dapat membantu para penyelidik, polis, peguam serta pihak berkuasa yang terlibat dalam bidang jenayah menyelesaikan jenayah dengan lebih cepat dan berkesan. Kata kunci: pengekstrakan maklumat; pengecaman entiti nama; Bahasa Melayu, berita jenayah, pendekatan berasaskan peraturan.

show abstract

Section: Latar Belakang Kajianunclassified

Pendekatan Teknik Pengecaman Entiti Nama Bagi Capaian Berita Jenayah Bahasa Melayu (Named Entity Recognition Approach for Malay Crime News Retrieval)

Saad

Mansor

2018

GEMA

View full text Add to dashboard Cite

show abstract

“…Association rules aim to extract interesting correlations, frequent patterns, associations or casual structures among sets of items in the given databases. Association rules mining has wide applications [35][36][37][38] in the areas of medical diagnosis, web caching, query expansion in information retrieval, homeland security, inventory control, market and risk management, telecommunication networks and so on.…”

Section: A Association Rules Miningmentioning

confidence: 99%

NER for Hindi language using association rules

Jain

Yadav

Tayal³

2014

2014 International Conference on Data Mining and Intelligent Computing (ICDMIC)

View full text Add to dashboard Cite

In this paper, we propose a state-of-art association rule mining algorithm for Hindi NER. Association rules are one of the key components of the data mining. Mined rules are of -TYPE 1, TYPE 2 and Type 3 i.e. dictionary, bi-gram and feature rules respectively. We consider corpus of news articles (100 training and 50 test sets) from leading Hindi newspapers. Hindi NER shows significant increase in performance when TYPE 2 rules are combined with TYPE 1 or with TYPE 3.

show abstract

“…However, research on this area for Indonesian, such as spelling checker 2,3 , are very few. Researchers who work on this language are still struggling to develop language resources like name entity recognizers (NER), annotated corpora, morphological analyzers, or parsers 4,5,6,7,8 to explore more sophisticated methods that have been applied to other languages. The limitations are not only on the existence of mature language tools but unfortunately, also on the availability of the annotated language data.…”

Section: Introductionmentioning

confidence: 99%

Exploiting Syntactic Similarities for Preposition Error Corrections on Indonesian Sentences Written by Second Language Learner

Irmawati

Shindo

Matsumoto

2016

Procedia Computer Science

View full text Add to dashboard Cite

We propose a method to artificially generate training data to correct preposition errors in Indonesian sentences written by second language learners. Basically, we injected large size of native sentences with preposition errors learned from learners' sentences. Our method copies a preposition error from a learner sentence to a native sentence by firstly calculating a syntactic similarity score between the native sentence and the learners' sentence. Then, it chooses the preposition error from the learner sentence that has the highest syntactic similarity score to the native sentence, to replace the original preposition in the native sentence.Experimental results show that the preposition error correction model trained on the artificial data resulted from our method outperforms the correction model trained on the similar size of native data.

show abstract

Named Entity Recognition for the Indonesian Language: Combining Contextual, Morphological and Part-of-Speech Features into a Knowledge Engineering Approach

Cited by 28 publications

References 7 publications

Pendekatan Teknik Pengecaman Entiti Nama Bagi Capaian Berita Jenayah Bahasa Melayu (Named Entity Recognition Approach for Malay Crime News Retrieval)

Pendekatan Teknik Pengecaman Entiti Nama Bagi Capaian Berita Jenayah Bahasa Melayu (Named Entity Recognition Approach for Malay Crime News Retrieval)

NER for Hindi language using association rules

Exploiting Syntactic Similarities for Preposition Error Corrections on Indonesian Sentences Written by Second Language Learner

Contact Info

Product

Resources

About