Showing papers by "Manjira Sinha published in 2015"

Computational Modeling of Morphological Effects in Bangla Visual Word Recognition.

Abstract: In this paper we aim to model the organization and processing of Bangla polymorphemic words in the mental lexicon. Our objective is to determine whether the mental lexicon accesses a polymorphemic word as a whole or decomposes the word into its constituent morphemes and then recognize them accordingly. To address this issue, we adopted two different strategies. First, we conduct a masked priming experiment over native speakers. Analysis of reaction time (RT) and error rates indicates that in general, morphologically derived words are accessed via decomposition process. Next, based on the collected RT data we have developed a computational model that can explain the processing phenomena of the access and representation of Bangla derivationally suffixed words. In order to do so, we first explored the individual roles of different linguistic features of a Bangla morphologically complex word and observed that processing of Bangla morphologically complex words depends upon several factors like, the base and surface word frequency, suffix type/token ratio, suffix family size and suffix productivity. Accordingly, we have proposed different feature models. Finally, we combine these feature models together and came up with a new model that takes the advantage of the individual feature models and successfully explain the processing phenomena of most of the Bangla morphologically derived words. Our proposed model shows an accuracy of around 80 % which outperforms the other related frequency models.

Compositionality in Bangla Compound Verbs and their Processing in the Mental Lexicon

Abstract: We conduct a cross-modal priming experiment to determine the mental representation and access strategies for compound verbs (CV) in Bangla. Analysis of reaction time indicates that compositionality among CVs triggers priming effects for both the constituent verbs. On the other hand non-compositional CVs exhibit priming only for the polar verb. Thus, compositional CVs are decomposed into their constituent verbs during processing. On the other hand, non-compositional verb phrases are represented and accessed as a whole in the minds of a Bangla speaker. The reaction time data thus collected are used to evaluate our vector space model for compositionality judgment.

Development of information processing, and communication applications for blind people in India

Abstract: BACKGROUND: Braille and text to speech systems have tremendously improved the lives of the visually impaired across the world. However, a large population of the visually impaired people in India is still deprived of getting the benefits of such advancements. This was primarily due to the difference in the technology required for Indian languages compared to those corresponding to other popular languages of the world. OBJECTIVE: In this paper, we present a design and evaluation of a unified open-source framework that will enable a visually impaired to access and produce information contents in Indian vernaculars with the help of computers. METHODS: The proposed framework has been designed by keeping constant interaction with the visually impaired people and by keeping in mind their different requirements. The framework is integrated with five different engines namely, a forward and reverse Indian language text to Braille transliteration engine, a Nemmeth editor to convert mathematical expressions into Braille, a tactile converter to convert graphical images into Braille, a voice mailing system to compose and receive voice based emails for the Blinds, and an Indian language file reader with integrated Hindi and Bangla speech synthesis engines. RESULTS: The framework has been deployed and exhaustively evaluated by people from different Blind schools and organizations for Blinds all over India. The transliteration engines are particularly useful for creating a large volume of Indian language Braille books.