Malayalam Language Research Articles

Verb Phrases Alignment Technique for English-Malayalam Parallel Corpus in Statistical Machine Translation Special issue on MTIL 2017

Abstract Machine translation (MT) from English to foreign languages is a fast developing area of research, and various techniques of translation are discussed in the literature. However, translation from English to Malayalam, a Dravidian language, is still in the rising stage, and works in this field have not flourished to a great extent, so far. The main reason of this shortcoming is the non-availability of linguistic resources and translation tools in the Malayalam language. A parallel corpus with alignment is one of such resources that are essential for a machine translator system. This paper focuses on a technique that enables automatic setting up of a verb-aligned parallel corpus by exploring the internal structure of the English and Malayalam language, which in turn facilitates the task of machine translation from English to Malayalam.

Automatic syllabification of speech signal using short time energy and vowel onset points

This paper describes a language independent method for automatic syllabification of speech signal. This method utilizes the valleys in short time energy (STE) contour and location of vowel onset points (VOP) for marking the syllable boundaries. In the proposed method, automatic syllabification is performed in three steps. First, long silence/pause regions are marked with the help of speech/non-speech detection. Then VOPs are located from the Hilbert Envelope of LP residual. The existence of more than one VOP in a continuous speech region (identified using speech/non-speech detection in the first step) is an indication of syllable boundaries within the region. Location with minimum energy in the STE contour between two consecutive VOP is identified as the syllable boundary. Since automatic VOP detection algorithm fails to detect some of the VOPs, certain syllable boundaries will be missed. Therefore, at the third step, additional syllable boundaries are detected from STE contour by fixing a valley threshold which is equal to the mean value of STE corresponding to each speech region between two consecutive syllable boundaries. This method is evaluated for 50 sentences each in read, extempore and conversational mode speech of Malayalam and Bengali languages. Overall accuracy of 80% is obtained with ± 50 ms tolerance with reference to manually marked syllable boundaries for this database. Method also shows good accuracy in case of TIMIT and NTIMIT data without tuning of thresholds and other parameters. This method is useful for applications that do not require exact syllable boundaries, rather a meaningful separation of syllables. Application of this technique for prosody based emotion recognition is illustrated using Emo-DB German emotional database.

A Comparative Study of Malayalam Literature and Paintings: Trajectories of Evolution

Art and literature are part as well as a reflection of life. Literature and arts help to observe and interpret the world. They can also change the world. Visual arts stand in the forefront of knowledge dissemination. However the significance accorded to literature has never been given to painting. The paper traces how literature has always preceded and given more priority against painting and other visual arts form in the region of Kerala. This also goes with the fact that in discussing the history of arts, rural arts and artists are never discussed sufficiently. This is despite the fact that there is always a closer relationship between alphabets, scripts and paintings of various forms. The paper traces this close relationship to the earlier times when the scripts and written forms essentially evolved from hand drawings and stone carvings in the context of the south Indian language Malayalam. That the scripts and alphabets essentially evolved from those early pictographs should be a useful background to understand the relationship. The paper also marks the historical transitions in the Malayalam alphabets and scripts under various influences.Keywords: Evolution of Malayalam Script, Vamozhi

Deep learning based spell checker for Malayalam language

Deep learning based spell checker for Malayalam language

Using effective web-based tools to address cancer health care disparities.

76 Background: Most of the World’s cancer patients live in non-English speaking world. However, there is a lack of authentic web-based cancer informational tools in non-English language, especially in low and middle-income countries. With increasing access to internet, there is an emergent need to democratize health information in native languages. The senior author was sensitive to this unmet need and created a comprehensive cancer website in Malayalam, a language spoken by more than 35 million people (www.keralacancercare.com). We describe the steps in creating the web resource and provide data on its use across various web and social media platforms. Methods: The senior author produced educational video sessions and cancer literature in Malayalam language. The website had three main sections – cancer basics, cancer A-Z and cancer FAQs. Cancer basics section contained sections on various basic cancer information such as types of malignancies, definitions, staging, symptoms, risk factors, diagnostic tests and screening. It also included sections on various treatment modalities, treatment side effects and information on cancer risk reduction, genetics and prevention. Cancer A-Z section contained instructional videos on major cancers. Cancer FAQs addressed common cancer myths and questions. A social media profile along with contact information through email service was also launched. The total cost incurred for development of such a web-based tool with presence on social media was less than USD 1500. Results: Since the launch of the web resource in May 2017, the website has now been accessed by more than 5000 unique visitors from 84 countries. More than 50 cancer related questions have been addressed since the launch. The web views of some of the brief informational videos such as ‘how to reduce risk for cancer’ has been viewed more than 150,000 times in social media pages. Conclusions: Creating authentic cancer informational web-based tools will have an immediate impact on addressing cancer healthcare disparities around the world. National and regional cancer societies must foster and encourage creation of novel healthcare informational tools that can truly democratize and empower people.

History of focus-concord constructions and focus-associated particles in Sinhala, with comparison to Dravidian and Japanese

This study traces the historical development of the focus concord construction of Sinhala from the language of the pre-second millennial graffiti on the Mirror Wall at Sigiriya to the modern colloquial language, with comparison to the historical development of focus concord constructions in the south Dravidian languages Malayalam and Tamil, as well as the focus concord (kakari-musubi) construction of Japanese. I argue that the Sinhala focus concord construction originated as one particular usage of impersonal verbal nominalisations in Old Sinhala, developed into a predicative clefting construction in Classical Sinhala, and in the modern colloquial language has become a phenomenon involving verb forms showing a sort of agreement with focussed elements.

Preparation, Validation and User-testing of Patient Information Leaflets on Diabetes and Hypertension

A patient information leaflet is an educational material for educating patients regarding a disease, medications and lifestyle modifications for better care. The present study aimed at preparing, validating and user-testing of diabetes and hypertension information leaflets in different languages among patients. The patient information leaflets on diabetes and hypertension were prepared by referring to the primary, secondary and tertiary resources. The content of the leaflet was validated by an expert committee. The readability of leaflets has been assessed by Flesch Reading Ease and Flesch-Kincaid Grade Level scores using MS Office 2007. The patient information leaflets were translated into Kannada and Malayalam languages with the help of language experts. During the user-testing, baseline knowledge was assessed before providing the patient information leaflet followed by provision of leaflets to the patients. After allowing the patients to read the leaflet for a period of 20 min, patients have again been administered with a set of questions to assess their knowledge. The Flesch Reading Ease and Flesch-Kincaid Grade Level readability scores achieved for diabetes leaflets are 70.4 and Flesch-Kincaid Grade Level score achieved is 6.3 and for hypertension leaflets, the Flesch Reading Ease score is 68.3 and Flesch-Kincaid Grade Level score achieved is 6.9. User testing of the patient information leaflets on diabetes and hypertension in different languages have been carried out on 40 each diabetic and hypertensive patients using knowledge based questionnaire. Knowledge assessment after providing patient information leaflet, the mean scores have significantly risen from 44.80 to 87.99 and 63.19 to 90.13 with p<0.0001, respectively in diabetes and hypertensive patients. It showed that patient education added with a well-designed patient information leaflet has a greater impact on knowledge of the patients towards their disease management, so that the medication adherence is improved.

Influence of Malayalam on temporal clauses in Malabar Indo-Portuguese

AbstractTemporal clauses with the subordinatorskandə(&lt; Portuguesequando) andk(w)ain Malabar Indo-Portuguese creole (MIP) are used to express any kind of temporal relation between two clauses, typically sequence or simultaneity. These temporal clauses are ubiquitous in contexts in which Portuguese, the lexifier of MIP, could not employ temporalquandoclauses. In this paper, I show that the morphosyntax and semantics of temporal clauses withkandəandk(w)ain MIP differ from corresponding Portuguese strategies, and that these differences can be explained by the influence of Malayalam (Dravidian), the substrate and adstrate language of MIP. One of the most salient properties of Malayalam adverbial subordination present in MIP is clause chaining. I position this study within the debate on creole exceptionalism, and show that the South Asian typological profile of MIP can only be explained within the view that language ecology determines the typology of a creole (Ansaldo 2009).

CULTURAL MINORITIES AND THE PANOPTIC GAZE: A STUDY OF THE (MIS)REPRESENTATION OF ETHNIC MINORITIES IN MALAYALAM FILMS

This paper explores the patterns of the representation of Adivasis or aboriginals – known as ‘tribals’ in common parlance – in Malayalam language films. Film as a medium of representation is continuously engaged in constructing images and thus the process becomes an ideological enterprise contributing to the relentless practice of defining and redefining the society and its various components in terms of several binaries. The film industry of Kerala, a southern state of India, is affluent and more influential than other art forms and production. Though the tribal population of Kerala is around 400 thousand and they belong to as many as 43 subgroups, they are underrepresented in films and that too is in a stereotypical manner. These groups are considered to be largely distinct with each tribal group identifying themselves with their own mythologies, tales of origin, and distinctive religious and ritualistic practices. This paper critically analyses the politics of representation using the example of tribals in Malayalam films as it has evolved over the past decades and attempts to trace a whole gamut of aesthetic and ethical issues at stake.

Screening for Depression among Community-dwelling Elders: Usefulness of the Center for Epidemiologic Studies Depression Scale.

Context:Though common, depressive disorders often remain undetected in late life.Aim:To examine the usefulness of Center for Epidemiologic Studies Depression (CES-D) for identifying depression among older people.Settings and Design:Community resident older people (aged 65 years or more), were evaluated by clinicians trained in psychiatry, as part of a cross-sectional study of late-life depression. Assessments were done in the community.Methods and Material:The participants were assigned ICD-10 diagnoses and assessed using Montgomery-Asberg Depression Rating Scale (MADRS) and CES-D. A short version of CES-D with 10 items, translated to the local language Malayalam, was used.Statistical Analysis:The sensitivity and specificity of CES-D was evaluated against ICD-10 clinical diagnosis of depression. The correlation of CES-D and MADRS was assessed using Pearson correlation coefficient.Results:220 consenting adults from 3 wards of the Panchayath were assessed. On analysis of the Receiver Operating Characteristic (ROC) curve of CES-D scores in relation to clinical diagnosis, the large Area Under Curve (AUC) showed efficient screening and a cut off score of 4 in CES-D had a sensitivity of 97.7% and a specificity of 79.1% for depression. There was also good correlation between the MADRS and CES-D scores (0.838).Conclusion:CES-D is a short simple scale which can be used by health care professionals for detecting depression in older people in primary care settings.

Frequency importance function for monosyllables in Malayalam

Objectives: The phonetic content of Malayalam is different from English and other Indian languages as it may contain more low frequency energy. The present study aimed to determine the frequency importance function (FIF) for monosyllables of Malayalam Language.Study design: Nineteen monosyllables spoken by a native female Malayalam speaker was used for the study. The speech stimuli were digitally filtered into 20 high pass and 20 low pass conditions. Speech noise was added to create nine conditions with SNRs of +8, +6, +4, +2, 0, −2, −4, −6 and −8 dB. The filtered stimuli were presented at these SNRs to 10 normal-hearing Malayalam speakers and were asked to identify the phonemes.Results: The FIF estimated in Malayalam language was compared with the FIF in English obtained from the published literature. The result shows that Malayalam has more weightage for low frequencies compared to English. The probable reason could be because of the inherent phonetic differences. The use of more nasalized speech which has low frequency content might have led to this result.Conclusions: The results of the study suggest that hearing aid fitting in Malayalam speakers should focus on providing more gain at low frequencies for better hearing aid benefit. The study needs to be performed on a larger population for better generalization of the results.

The “Worlds” Of the Region

The “region”—understood in terms of the linguistically organized states—has been one of the central categories around which debates about cultural politics in India have been organized in the last two decades. This article argues that within this scholarship, the “region” has figured as a mirror of the nation, either as continuing from the impulses of the nation or as a response to it. Contra this, it proposes that “region” should be understood as being formed simultaneously within and without the structuring of the nation. Even while the horizon of universality provided by the nation to its subjects presents the figure of the “citizen” as its teleology, nonnational resources are mobilized to configure the subject of the region. Rather than a historical development, this performative subject, who works with multiple horizons of universality, is foundational to the formation of the region. This article, using two Malayalam language films from the South Indian state of Kerala, Arabikatha (Arabian Tale) and 16 MM: Memories, Movement and a Machine, attempts to think through how mistranslation of the “world” is central to the formation of the regional subject.

Word Retrieval Ability on Phonemic Fluency in Typically Developing Children

Verbal fluency tasks are simple behavioral measures useful in assessing word retrieval abilities. Among the verbal fluency tasks, the utility of the Phonemic Fluency Task in children has received less attention. As the task is dependent on phonemic characteristics of each language, there is a great need for understanding its developmental trend. The present study, therefore, aims to delineate the performance on phonemic fluency in typically developing Malayalam-speaking children. Verbal fluency performance on 2 tasks of phonemic fluency was tested using a cross-sectional study design among 1,015 school-going Malayalam-speaking typically developing children aged 5 to 15 years old. Performance with respect to word productivity and clustering-switching measures was analyzed. The effect of age, gender, and tasks on the outcome measures were investigated in the present study. Study findings revealed a positive influence of age with no statistically significant gender effects. Children employed both task-discrepant and task-consistent organizational strategies during tasks of phonemic fluency, dependent purely on the Malayalam language. Future research focusing on developmental trends across different languages is vital for enhancing the task's clinical sensitivity and specificity among childhood disorders.

Annotations of Connectives and Arguments in Malayalam Language

Discourse relations in natural languages link clauses in text and compose overall text structure. Discourse connectives are an important part of modeling the Malayalam discourse structure. We followed the annotation procedure of Penn Discourse Tree Bank and worked on tagging of discourse connectives and arguments of Malayalam text and also report the senses of relation. We present our work on annotations of Malayalam discourse connectives and arguments which helps to know more about the discourse connectives and their appearance in case of semantic rules in Malayalam discourse. Discourse connectives may or may not be explicitly present in the relation. In our work, we focus on the annotation of both explicit and implicit connectives and arguments in Malayalam text and showed encouraging results.

Performance Analysis of Malayalam Language Speech Emotion Recognition System Using ANN/SVM

Automatic recognition of emotions from speech by machines has been one of the most challenging areas of research in the field of human machine interaction. Automatic emotion recognition system by speech merely means that to monitor and identify the emotional or physiological state of an individual from their utterances. Speech emotion recognition has wide range of application ranging from clinical studies to robotics. In this paper developed speech emotional database for Malayalam language (One of the south Indian languages) and a system for recognizing the emotions. The system used Mel Frequency Cepstral Coefficients (MFCCs), Short Time Energy (STE) and Pitch as features extraction techniques. Two classifiers, namely Artificial Neural Network (ANN) and Support Vector Machine (SVM) used for pattern classification. Experiments show that this method provides a high accuracy of 88.4% in the case of ANN and 78.2% in the case of SVM.

Comparative Study of Different Classifiers for Malayalam Dialect Recognition System

Malayalam is a South Indian language spoken predominantly in the state of Kerala. In this paper, a comparative study of different classifiers to recognize Malayalam language dialects has been carried out and presented. Thrissur and Kozhikode are the two different dialect corpora used for the recognition task. Mel Frequency Cepstral Coefficients (MFCC), energy and pitch are the features extracted. Then these feature vector set obtained are classified in the classification phase using Artificial Neural Networks (ANN), Support Vector Machine (SVM) and Naive Bayes classifiers. The input feature vector data is trained using data relating to patterns which are known and using the test data set they are tested further. Based on recognition accuracy, the performance of the ANN, SVM, and Naive Bayes are evaluated. Speech recognition is a multiclass classification problem. During classification stage, the input feature vector data is trained using information relating to known patterns and then they are tested using the test data set. ANN produced a recognition accuracy of 90.2%, SVM produced an accuracy of 88.2% and Naive Bayes produced an accuracy of 84.1%. Among the three classifiers, ANN is found to be better.

Malayalam Question Answering System

Question answering system, an important part in natural language processing aims at automatically finding concise answers to arbitrary questions phrased in natural language. The goal is difficult while considering the agglutinative nature of south Indian language Malayalam. Studies indicating that the usage of Malayalam documents on the web is increasing. In this paper we aim at retrieving factoid answers for the questions in Malayalam from a given set of Malayalam documents under a closed domain. TnT tagger is used to train the corpus of words inorder to find the precise factoid answers.

'SEEDS' of 'Good Lessons' through 'Many a Drop'-- Media Initiation in Environmental Education: An Indian Model of Environmental Pedagogy

Environmental communication is now an emerging and a significant curriculum from schools to research centers. The effective and efficient environmental communication occurs when learners interact with their surrounding environment/ecology in which they live and reciprocate for sustainable protection and restoration of it. Developing countries in Asia and Africa are now setting up new role models and practices in curricula of environmental communication. The traditional theory based environmental communication curriculum of the last century is now actively investigated and restructured through community based learning, affirmative actions, and student centered participatory curriculum. Kerala, a southern State in India, serves as an exemplar of this new eco-venture. Through case studies like, Nalla Paadam (Good Lesson), Palathulli Project (Many a Drop Project) by the Malayalam language daily ‘Malayala Manoram’, and SEED project by another Malayalam daily ‘Mathrubhumi’, this paper analyses the innovative curriculum practices in the state of Kerala in India.

Handwritten Malayalam Character Recognition using Curvelet Transform and ANN

Malayalam, the official language of Kerala, a southern state of India has been accorded the honour of language of eminence. Hence the researches in recognition and related works in Malayalam language is gaining more prominence in the current scenario. This paper proposes the use of Curvelet transform and neural network for the recognition of handwritten Malayalam character. Curvelet transform is to be used in the feature extraction stage and neural network for classification. Curvelet transform provides a compact representation for curved singularities and is well suited for malayalam language. Two different back propagation algorithms had been employed and the performance is compared on varying architecture. The promising feature of the work is successful classification of 53 characters which is an improvement over the existing works. Application of character recognition include sorting of bank cheques and postal letters, reading aid for blind, data compression etc. Besides, an automated tool with graphical user interface in MATLAB has been developed for Malayalam character recognition. General Terms Pattern Recognition, Artificial Neural Network (ANN), Curvelet Transform, Optical character recognition (OCR),

Thiruvithamkoor, Malabar, Kerala: Speculations on the Regions in “Regional Cinema”

The category of the region, in its use as an adjective to cinema, presumes a convergence of language, culture, and geography as its ground. This article attempts to understand the dynamics of unification and difference, aggregation and disaggregation that is central to the production of the linguistic region. Focusing on cinematic practices of the early 1950s in Malayalam, it examines the conditions under which a Thiruvithamkoor (Travancore)-based film industry transforms into an industry for “unified Kerala.” Sidestepping the teleological narrative of unification, the article attends to this period as one characterized by many possible futures. The article is organized around two sets of texts. The first set comprises Chalachitra Kala (The Art of Moving Pictures) by Nagavally R.S. Kurup and Cinima (Cinema) by Moorkoth Kunhappa. Published in 1951, these were the first books on cinema to be written in the Malayalam language. I read these books to understand the differential relationship to cinema that seemed to exist within the Malayalam-speaking territories—a difference founded on the ways in which the industry was organized, publics constituted, and markets formed. The second set comprises three films— Jeevitha Nauka (1951), Neelakkuyil (1954), and Thiramala (1953). These films demonstrate how the discourse of stardom during this crucial period, between the end of the colonial regime and linguistic reorganization of states in independent India, engendered industrial and narrative assemblages that (re)constituted the imaginaries of the region.