“A careless word may kindle strife,
A cruel word may wreck a life,
A timely word may lessen stress,
A loving word may heal and bless”

Unknown Author

Bali Ranaivo-Malançon loves WORDS and she ends up studying and teaching Natural Language Processing (NLP)!

Bali was born and raised in Madagascar (the fourth biggest island in the world!), moved to France, and then, in 2002, went to Malaysia to merantau (“wander about” by Google Translate). She received her Ph.D. in NLP from the National Institute for Oriental Languages and Civilizations (INALCO, Paris, France) in 2001. She is currently an Associate Professor at the Universiti Malaysia Sarawak, located in Borneo, the third biggest island in the world.

Bali’s research interests are geared towards the development of linguistic resources (collection, formalisation, and annotation), text processing, and text mining. She has already developed a few text processing tools for Malay language. She created a recogniser of foreign words in Malay texts during her D.E.A research. The recognition is done through a list of patterns that she obtained after studying deeply Malay morphographemic structures. The output of her Ph.D. research is an analyser of Malay affixed words. She has determined the morphological structures of Malay words, created rules for automatic analysis of these words, designed the architecture of the morphological analyser, and then implemented it. She has also designed a language identifier, which can recognise the language of short segments of a text and able to distinguish Malay texts from Indonesian texts (or vice-versa). The algorithm is based on trigrams and non-shared words. She has also designed the first Malay corpus processing tool named CorPlus. The implementation in Java was done by her final year project students in 2005.  

During her ten years in Malaysia, she had the opportunities to work as a consultant for Malaysian companies and government institutions. She collaborated also in a few international projects.

Bali is now focusing on the design of a “pipeline” method to speed up the processing of Sarawak Indigenous languages and on the development of a digital repository to host Sarawak historical documents.