Code-mixing and code-switching (CMCS) are frequent features in online conversations. Classification of such text is challenging if one of the languages is low-resourced. Fine-tuning pre-trained multilingual language models (PMLMs) is a promising avenue for code-mixed text classification. In this paper, we explore adapter-based fine-tuning of PMLMs for CMCS text classification. We introduce sequential and parallel stacking of adapters, continuous fine-tuning of adapters, and training adapters without freezing the original model as novel techniques with respect to single-task CMCS text classification. We also present a newly annotated dataset for the classification of Sinhala-English codemixed and code-switched text data, where Sinhala is a low-resourced language. Our dataset of 10000 user comments has been manually annotated for five classification tasks: sentiment analysis, humor detection, hate speech detection, language identification, and aspects identification, thus making it the first publicly available Sinhala-English CMCS dataset with the largest number of annotation types. In addition to Adapter Based Fine-Tuning of PMLMs for CMCS Text Classification this dataset, we also carried out experiments on our proposed techniques with Kannada-English and Hindi-English datasets. These experiments confirm that our adapter-based PMLM fine-tuning techniques outperform, or are on par with the basic fine-tuning of PMLM models.

The Department of Information Science is one of seven departments that make up the School of Business at the University of Otago. The department offers courses of study leading to a major in Information Science within the BCom, BA and BSc degrees. In addition to undergraduate teaching, the department is also strongly involved in postgraduate research programmes leading to MCom, MA, MSc and PhD degrees. Research projects in spatial information processing, connectionist-based information systems, software engineering and software development, information engineering and database, software metrics, distributed information systems, multimedia information systems and information systems security are particularly well supported.The views expressed in this paper are not necessarily those of the department as a whole. The accuracy of the information presented in this paper is the sole responsibility of the authors. CopyrightCopyright remains with the authors. Permission to copy for research or teaching purposes is granted on the condition that the authors and the Series are given due acknowledgment. Reproduction in any form for purposes other than research or teaching is forbidden unless prior written permission has been obtained from the authors. CorrespondenceThis paper represents work to date and may not necessarily form the basis for the authors' final conclusions relating to this topic. It is likely, however, that the paper will appear in some form in a journal or in conference proceedings in the near future. The authors would be pleased to receive correspondence in connection with any of the issues raised in this paper, or for subsequent publication details. Please write directly to the authors at the address provided below. (Details of final journal/conference publication venues for these papers are also provided on the Department's publications web pages: http://www.otago.ac.nz/informationscience/pubs/). Any other correspondence concerning the Series should be sent to the DPS Coordinator. Abstract. Second Life is a multi-purpose online virtual world that provides a rich platform for remote human interaction. It is increasingly being used as a simulation platform to model complex human interactions in diverse areas, as well as to simulate multi-agent systems. It would therefore be beneficial to provide techniques allowing high-level agent development tools, especially cognitive agent platforms such as belief-desire-intention (BDI) programming frameworks, to be interfaced to Second Life. This is not a trivial task as it involves mapping potentially unreliable sensor readings from complex Second Life simulations to a domain-specific abstract logical model of observed properties and/or events. This paper investigates this problem in the context of agent interactions in a multi-agent system simulated in Second Life. We present a framework which facilitates the connection of any multi-agent platform with Second Life, and demonstrate it in conjunction with an extension of the Jason BDI interpreter. Department of Inform...

This paper addresses the issue of integrating agents with a variety of external resources and services, as found in enterprise computing environments. We propose an approach for interfacing agents and existing message routing and mediation engines based on the endpoint concept from the enterprise integration patterns of Hohpe and Woolf. A design for agent endpoints is presented, and an architecture for connecting the Jason agent platform to the Apache Camel enterprise integration framework using this type of endpoint is described. The approach is illustrated by means of a business process use case, and a number of Camel routes are presented. These demonstrate the benefits of interfacing agents to external services via a specialised message routing tool that supports enterprise integration patterns.

Second Life is one of the most popular multipurpose online virtual worlds, which supports applications in diversified areas relating to real-life activities. Moreover, it is possible to use Second Life in testing Artificial Intelligence theories, by creating intelligent virtual agents. For the successful implementation of many of these applications, it is important to accurately identify events taking place inside Second Life. This involves extracting low-level spatio-temporal data and identifying the embedded high-level domain-specific information. This is an aspect that has not been taken into consideration in the prior research related to Second Life. This paper presents a framework that extracts data from Second Life with high accuracy and high frequency, and identifies the high-level domain-specific events and other contextual information embedded in these low-level data. This is guided by our virtual environment formalism, which defines events and states in a virtual environment. This framework is further enhanced to be connected with multi-agent development platforms, thus demonstrating its use in the area of Artificial Intelligence.

Neural Machine Translation (NMT) has seen a tremendous spurt of growth in the last twenty years and has already entered a mature phase. While considered the most widely used solution for Machine Translation, its performance on low-resource language pairs remains sub-optimal compared to the high-resource counterparts due to the unavailability of large parallel corpora. Therefore, the implementation of NMT techniques for low-resource language pairs has been receiving the spotlight recently, thus leading to substantial research on this topic. This paper presents a detailed survey of research advancements in low-resource language NMT (LRL-NMT) and quantitative analysis to identify the most popular techniques. We provide guidelines to select the possible NMT technique for a given LRL data setting based on our findings. We also present a holistic view of the LRL-NMT research landscape and provide recommendations to enhance the research efforts further.

Today, comprehending consumer behavior is becoming dynamically challenging with the emergence of social commerce. Business organizations are now striving to convince consumers by exploiting the advantage of social support empowered by online social networks. Importantly, social ties in such online social networks facilitate trust as the most compelling benefit while alleviating the perceived risk, which happened to be the major concerns with electronic commerce over the years. This research study is aimed at understanding the impact of social commerce on the consumer behavior, particularly consumer decision-making stages. Hence, this research was conducted as a quantitative study involving a cross-sectional survey and gathered valid responses from Facebook users. Structural Equation Modeling (SEM) was used to analyze data and test hypotheses. The findings exhibited significant positive effects from social commerce on all the consumer decision-making stages namely; need recognition, information search, alternative evaluation, purchase decision and post-purchase decision. Therefore, this study highlights the importance of employing an appropriate social commerce strategy for business organizations.

Current state-of-the-art speech-based user interfaces use data intense methodologies to recognize free-form speech commands. However, this is not viable for low-resource languages, which lack speech data. This restricts the usability of such interfaces to a limited number of languages. In this paper, we propose a methodology to develop a robust domainspecific speech command classification system for low-resource languages using speech data of a high-resource language. In this transfer learning-based approach, we used a Convolution Neural Network (CNN) to identify a fixed set of intents using an ASR-based character probability map. We were able to achieve significant results for Sinhala and Tamil datasets using an English based ASR, which attests the robustness of the proposed approach. .

No abstract