Last year we conducted research to explore the possibility of leveraging online job portal data for economic analysis in 13 Asia Pacific countries, as a part of a project for the Asian Development Bank. We examined the types of information available on major portals across the region, to discern the nature and format of available data. We also tested and refined methodologies to analyse a dataset comprising online job vacancies sourced from a Sri Lankan job portal, to demonstrate use cases for exploring the impacts of shocks on the labour market. The first step in this exploration was to review where in practice online job portal data has been used, to identify the methods and techniques available along with their strengths and limitations. The full review is published below. 
Last year we conducted research to explore the possibility of leveraging online job portal data for economic analysis in 13 Asia Pacific countries, as a part of a project for the Asian Development Bank. We examined the types of information available on major portals across the region, to discern the nature and format of available data. We also tested and refined methodologies to analyse a dataset comprising online job vacancies sourced from a Sri Lankan job portal, to demonstrate use cases for exploring the impacts of shocks on the labour market. The first step in this exploration was to review where in practice online job portal data has been used, to identify the methods and techniques available along with their strengths and limitations. The full review is published below. 
On 2nd October 2023, Research Manager and Team Lead (Data, Algorithms, and Policy) Merl Chandana, alongside Junior Researcher Chanuka Algama, held a session titled ‘Applied data science research for social good’ at the University of Kelaniya’s Department of Statistics and Computer Science. The session delved into LIRNEasia’s journey of forming a data science team and using large datasets to yield critical insights for public policy. They contrasted LIRNEasia’s applied data science approach with traditional academic research and private sector practices. Additionally, they highlighted the emerging ‘AI for Social Good’ movement and its potential as a career avenue. The slides used can be accessed below.
By employing unsupervised and supervised machine learning techniques, we explore the feasibility of utilizing mobile call detail records (CDRs) as well as geographic information system (GIS) and remote sensing (RS) data to map poverty spatially
Many countries around the world have adopted artificial intelligence (AI) polices. However, Sri Lanka is yet to adopt one. This discussion paper considers factors that may be taken into account if an AI policy were to be drafted in Sri Lanka.
This policy brief looks at the current status of Sri Lanka's Open Data Portal, and what may be done to improve it.
Keynote presentation for South Eastern University, 10th Annual Science Research Sessions 2021, 30 November 2021 - by Rohan Samarajiva, LIRNEasia
Over the past decade, both internet penetration and digital media user base have increased substantially.
We present a dataset consisting of 3468 documents in Bengali, drawn from Bangladeshi news websites and factchecking operations, annotated as CREDIBLE, FALSE, PARTIAL or UN-CERTAIN. The dataset has markers for the content of the document, the classification, the web domain from which each document was retrieved, and the date on which the document was published. We also present the results of misinformation classification models built for the Bengali language, as well as comparisons to prior work in English and Sinhala.
This research report analyses the implementation of AI ethics principles in the policy, legal and regulatory, and technical arenas in Singapore and India.
The Institute of Chartered Professional Managers of Sri Lanka’s (CPM Sri Lanka) 26th Webinar was held on the 27th of August 2021 with a focus on the ‘Safety of information in a technically driven world’ a timely subject of cyber security.
LIRNEasia Chair Prof. Rohan Samarajiva, shared his expertise on the key presentation (below) focusing on the current issues of information security, including potential risks to organizations and its management, data storage and back-ups as well as prevention and recovery.
We present a dataset consisting of 3576 documents in Sinhala, drawn from Sri Lankan news websites and factchecking operations, annotated as CREDIBLE, FALSE, PARTIAL or UN- CERTAIN. The dataset has markers for the content of the document, the classification, the web domain from which each document was retrieved, and the date on which the document was published. We also present the results of misinformation classification models built for the Sinhala language, as well as comparisons to English benchmarks, and suggest that for smaller media ecosystems it may make more practical sense to model uncertainty instead of truth vs falsehood binaries.
As hate speech on social media becomes an ever-increasing problem, policymakers may look to more authoritarian measures for policing content. Several countries have already, at some stage, banned networks such as Facebook and Twitter (Liebelson, 2017).
This paper presents two colloquial Sinhala language corpora from the language efforts of the Data, Analysis and Policy team of LIRNEasia, as well as a list of algorithmically derived stopwords. The larger of the two corpora spans 2010 to 2020 and contains 28,825,820 to 29,549,672 words of multilingual text posted by 533 Sri Lankan Facebook pages, including politics, media, celebrities, and other categories; the smaller corpus amounts to 5,402,76 words of only Sinhala text extracted from the larger.
We summarize the state of progress in artificial intelligence as used for classifying misinforma- tion, or ’fake news’. Making a case for AI in an assistive capacity for factchecking, we briefly examine the history of the field, divide current work into ’classical machine learning’ and ’deep learning’, and for both, examine the work that has led to certain algorithms becoming the de facto standards for this type of text classification task.
In a practical experiment, we benchmark five common text classification algorithms - Naive Bayes, Logistic Regression, Support Vector Machines, Random Forests, and eXtreme Gradient Boosting - on multiple misinformation datasets, accounting for both data-rich and data-poor environments.
Intended for policymakers, technologists, educators and others, this international collection of 19 short stories delves into AI’s cultural impacts with hesitation and wonder.