Browsing by Author "Panchendrarajan, R"

Now showing 1 - 13 of 13

Embargo
AI-Based Child Care Parental Control System
(IEEE, 2022-12-09) Jayasekara, U; Maniyangama, H; Vithana, K; Weerasinghe, T; Wijekoon, J; Panchendrarajan, R
Due to the prevalence of the COVID-19 epidemic around the globe, children were compelled to engage in remote learning through online platforms, hence mobile phone has become one of their predominant devices. Mobile device with Internet access offers a major outlet for education, entertainment, and social connection, but this combination can lead to several significant bad sequences such as online exploitation, harmful addictions, and other negative impacts of online social networking. To address harmful effects, parental controls are becoming more crucial, yet Sri Lankan parents are less aware of this. Consequently, this study proposes a parental control system to monitor their child’s activities. Android, Microsoft Azure, Java, Python, OpenCV, MySQL, and FastAPI are among the most prominent technologies utilized in the proposed application’s development. The suggested approach focuses primarily on the Sri Lankan context and aims to enhance parental digital literacy while safeguarding children from cyber threats. Yielded results showed the proposed mobile application for the identification of toxic words, drugs & alcohol content, game character images, and Instagram Sinhala comments severity as 94%, 95%, 97%, and 55% respectively in controlled experiments.
Open Access
Bidirectional LSTM-CRF for Named Entity Recognition
(32nd Pacific Asia Conference on Language, Information and Computation, 2018-12-01) Panchendrarajan, R; Amaresan, A
Named Entity Recognition (NER) is a challenging sequence labeling task which requires a deep understanding of the orthographic and distributional representation of words. In this paper, we propose a novel neural architecture that benefits from word and character level information and dependencies across adjacent labels. This model includes bidirectional LSTM (BI-LSTM) with a bidirectional Conditional Random Field (BI-CRF) layer. Our work is the first to experiment BI-CRF in neural architectures for sequence labeling task. We show that CRF can be extended to capture the dependencies between labels in both right and left directions of the sequence. This variation of CRF is referred to as BI-CRF and our results show that BI-CRF improves the performance of the NER model compare to an unidirectional CRF and backward CRF is capable of capturing most difficult entities compare to the forward CRF. Our system is competitive on the CoNLL-2003 dataset for English and outperforms most of the existing approaches which do not use any external labeled data.
Embargo
Characterisation of Mental Health Conditions in Social Media Using Deep Learning Techniques
(Springer, Cham, 2022-09-19) Sharma, T; Panchendrarajan, R; Saxena, A
Social media has become the easiest and most popular choice for online users to share their views, thoughts, opinions, and emotions with their friends and followers. The shared content is very helpful in making valuable conclusions about an individual’s personality. Nowadays, researchers across the world are collecting social media data for understanding the mental health of social media users. Mental illness is a big concern in today’s society, and in no time, it can turn into suicidal thoughts if efficacious methods are not taken into account. Early detection of such mental health conditions provides a potential way for effective social intervention. This has opened up opportunities to the research community to automatically determine various mental health conditions, such as anxiety, depression, and near-suicidal thoughts from user-generated content. The deep learning techniques have been extensively used in this research area to better understand the mental health of users on social networking platforms. In this chapter, we conduct a comprehensive review of the research works that use deep learning techniques to identify various mental health conditions from social media data. We discuss methods to detect different types of mental health conditions, such as anxiety, depression, stress, suicide, and anorexia. We also provide the details of the datasets used in these studies. The chapter is concluded with promising future directions.
Embargo
Cheap food or friendly staff? Weighting hierarchical aspects in the restaurant domain
(IEEE, 2016-04-05) Panchendrarajan, R; Murugaiah, B; Prakhash, S; Ahamed, N; Ranathunga, S; Pemasiri, A
In aspect-level opinion mining, each aspect is assigned a rating based on customer reviews. More often than not, these aspects exhibit a hierarchical relationship, and the restaurant domain is no difference. With the existence of such hierarchical relationships, rating of an aspect is based on the composite score of its sub-elements. However, the influence of these sub-aspects on the score of a parent aspect is not uniform, since some sub-aspects are perceived more important than others. Therefore, when calculating the composite score for an aspect, influence of each sub-aspect should be weighted according to its perceived importance. Identifying weights for different aspects is addressed as the problem of multi-attribute weighting. However the existing approaches do not utilize the relationships between aspects to find weights. This paper presents an approach to find weights for aspects that exhibit hierarchical relationships in restaurant domain using an improved version of the Analytic Hierarchy Process (AHP), one of the Multi Attribute Decision Making Techniques (MADTs). Different aspects of the restaurant domain are modeled as a hierarchy and weights for aspects are calculated using AHP. Occurrence counts of aspects in restaurant reviews are used to obtain the relative importance of aspects. This approach provides acceptable consistency ratios for the pairwise comparison matrices obtained for each level in the hierarchy of aspects.
Embargo
Cricket Shot Image Classification Using Random Forest
(IEEE, 2021) Devanandan, M; Rasaratnam, V; Anbalagan, M. K; Asokan, N; Panchendrarajan, R; Tharmaseelan, J
Cricket is one of the top 10 most played sport across the world regardless of age and gender. However, learning cricket has been quite challenging as the majority of the cricket-playing individuals are unable to afford quality infrastructure. While this has opened up many research opportunities to provide solutions to automatically learn cricket, very little work has been done in this era. In this paper, we focus on the batting skills of cricket players. We develop a Random Forest model to classify the cricket shot images using human body keypoints extracted with MediaPipe. Experiment results show the proposed model achieves an F1-score of 87% and outperforms the existing solution in a 5% margin. Further, we propose a similarity estimation approach to compare the user’s cricket image with popular international cricket players’ cricket shot images of the same type and retrieve the most similar one. The mobile application we developed based on our solution will enable cricket-playing individuals to analyze, improve and track their batting performances without the need of having a coach.
Open Access
Dataset Reconstruction Attack against Language Models
(2021-07) Panchendrarajan, R; Bhoi, S
With the advances of deep learning techniques in Natural Language Processing, the last few years have witnessed releases of powerful language models such as BERT and GPT-2. However, applying these general-purpose language models to domain-specific applications requires further fine-tuning using domain-specific private data. Since private data is mostly confidential, information that can be extracted by an adversary with access to the models can lead to serious privacy risks. The majority of privacy attacks on language models infer either targeted information or a few instances from the training dataset. However, inferring the whole training dataset has not been explored in depth which poses far greater risks than disclosure of some instances or partial information of the training data. In this work, we propose a novel data reconstruction attack that also infers the informative words present in the private dataset. Experiment results show that an adversary with black-box query access to a fine-tuned language model can infer the informative words with an accuracy of about 75% and can reconstruct nearly 46.67% of the sentences in the private dataset.
Open Access
Eatery: a multi-aspect restaurant rating system
(Association for Computing Machinery, 2017-07-04) Panchendrarajan, R; Ahamed, N; Sivakumar, p; Murugaiah, B; Ranathunga, S; Pemasiri, A
This paper presents Eatery, a multi-aspect restaurant rating system that identifies rating values for different aspects of a restaurant by means of aspect-level sentiment analysis. Eatery uses a hierarchical taxonomy that represents relationships between various aspects of the restaurant domain that enables finding the sentiment score of an aspect as a composite sentiment score of its sub-aspects. The system consists of a word co-occurrence based technique to identify multiple implicit aspects appearing in a sentence of a review. An improved version of Analytic Hierarchy Process (AHP) is used to obtain weights specific to a restaurant by utilizing the relationships between aspects, which allows finding the composite sentiment score for each aspect in the taxonomy. The system also has the ability to rate individual food items and food categories. An improved version of Single Pass Partition Method (SPPM) is used to categorise food names to obtain food categories.
Open Access
Emotion-Aware Event Summarization in Microblogs
(Association for Computing Machinery, 2021-04-19) Panchendrarajan, R; Hsu, W; Lee, M. L
Microblogs have become the preferred means of communication for people to share information and feelings, especially for fast evolving events. Understanding the emotional reactions of people allows decision makers to formulate policies that are likely to be more well-received by the public and hence better accepted especially during policy implementation. However, uncovering the topics and emotions related to an event over time is a challenge due to the short and noisy nature of microblogs. This work proposes a weakly supervised learning approach to learn coherent topics and the corresponding emotional reactions as an event unfolds. We summarize the event by giving the representative microblogs and the emotion distributions associated with the topics over time. Experiments on multiple real-world event datasets demonstrate the effectiveness of the proposed approach over existing solutions.
Embargo
iMask: An IoT-based Intelligent Mask to Identify and Track COVID-19 Suspects
(IEEE, 2022-09-08) Yamasinghe, N; Ranasinghe, Y; Dissanayake, Y; Wijekoon, J.L; Panchendrarajan, R
COVID-19 has become a global health concern, and wearing masks is a key measure to curb COVID-19 from rapidly spreading. While COVID-19 patients can be accurately determined using Rapid Antigen and PCR tests, these tests are costly, time-consuming, invasive, and uncomfortable. Further, they should be performed in a specialized environment despite showing the COVID-19 symptoms such as fever, cough, rapid heart rate, shortness of breath, and low blood oxygen saturation level. To this end, this study aims to automatically identify, and track the COVID-19 suspects in real-time by embedding smart sensors to face masks. The mask was developed to gather the data related to five major symptoms of COVID-19: body temperature, cough, heart rate, breathing pattern, and blood oxygen level. Data collected using smart sensors were used to identify and track COVID-19 suspects using Deep Neural Networks, the Internet of Things (IoT), and Artificial Intelligence (AI). Yielded results showed the proposed mask can identify COVID-19 suspects 92% accurately.
Open Access
Implicit aspect detection in restaurant reviews using cooccurence of words
(Proceedings of the 7th Workshop on computational approaches to subjectivity, sentiment and social media analysis, 2016-06) Panchendrarajan, R; Ahamed, N; Murugaiah, B; Sivakumar, P; Ranathunga, S; Pemasiri, A
For aspect-level sentiment analysis, the important first step is to identify the aspects and their associated entities present in customer reviews. Aspects can be either explicit or implicit, where the identification of the latter is more difficult. For restaurant reviews, this difficulty is escalated due to the vast number of entities and aspects present in reviews. The problem of implicit aspect identification has been studied for customer reviews in different domains, including restaurant reviews. However, the existing work for implicit aspect identification in customer reviews has the limitation of choosing at most one implicit aspect for each sentence. Furthermore, they deal only with a limited set of aspects related to a particular domain, thus have not faced the problem of ambiguity that arises when an opinion word is used to describe different aspects. This paper presents a novel approach for implicit aspect detection, which overcomes these two limitations. Our approach yields an F1- measure of 0.842 when applied for a set of restaurant reviews collected from Yelp.
Embargo
Tamil Grammarly – A Typing Assistant for Tamil Language using Natural Language Processing
(IEEE, 2023-06-12) Mahadevan, P; Srihari, P; Seyon, S; Vasavan, P; Panchendrarajan, R
Tamil is one of the ancient and most convoluted languages in the world. Although it is being the official language of many Asian countries, even native speakers tend to find difficulties in writing Tamil due to its morphologically rich nature. While there are various studies focusing on automatically identifying and correcting a specific typing error, very limited effort has been made to develop a comprehensive solution to assist the native and non-native writers of Tamil. In this paper, we propose a typing assistant tool Tamil Grammarly using Natural Language Processing (NLP) techniques. Specifically, the tool aims to aid the user to fix grammatical errors and spelling errors and recommend the next words and synonyms of the current word in real-time while typing. The NLP-based typing assistant functions of Tamil Grammarly were developed using a transformer-based model, LSTM model, and Word2Vec model. Extensive evaluation performed shows that our tool can assist the users in real-time with an accuracy of 73% - 93% within 0.4 to 5.3 seconds.
Embargo
tAssessee: Automatically Assessing Quality of Tea Leaves using Image Processing Techniques
(IEEE, 2022-11-30) Sivalingam, J; Sivachandrabose, L.N; Loganathan, M; Sivakumaran, J; Panchendrarajan, R
Sri Lanka is one of the well-known international’s pinnacle tea exporters with a high global demand attracting millions of foreign exchanges, which strengthens the economy of the country. Despite the fact that tea brings a good source of foreign exchange, the tea industry lacks efficiency and effectiveness during the assessment of plucked tea leaves which compromises the significant quality of tea. While studies have revealed various factors affecting the tea quality, key factors are identified as the presence of tea diseases, pest attacks, the mixture of fresh and mature tea leaves, and the mixture of tea grades present in the tea sack. In this paper, we focus on automatically assessing the quality of tea leaves for a single tea leaf and bulk tea leaves before initiating the tea manufacturing process. The proposed tAssessee system allows the user to upload the image of a single tea leaf or bulk tea leaves to automatically assess four different quality factors of tea leaves such as disease, pest attack, freshness, and grade using Convolutional Neural Network based models and using various image processing techniques. This will assist the tea supervisors in the tea factories to automatically assess the quality of tea leaves where the manufacturing process can be segregated according to the quality of tea leaves and determine the pricing accordingly. Extensive experiments performed using the tea leaves images gathered in tea factories reveal, that the proposed tAssessee system can assess the quality of single tea leaf and bulk tea leaves with the accuracy range of 87% - 98% and 91% - 100% respectively.
Open Access
Topic-based influential user detection: a survey
(Springer, Cham, 2022-07-05) Panchendrarajan, R; Saxena, A
Online Social networks have become an easy means of communication for users to share their opinion on various topics, including breaking news, public events, and products. The content posted by a user can influence or affect other users, and the users who could influence or affect a high number of users are called influential users. Identifying such influential users has a wide range of applications in the field of marketing, including product advertisement, recommendation, and brand evaluation. However, the users’ influence varies in different topics, and hence a tremendous interest has been shown towards identifying topic-based influential users over the past few years. Topic-level information in the content posted by the users can be used in various stages of the topic-based influential user detection (IUD) problem, including data gathering, construction of influence network, quantifying the influence between two users, and analyzing the impact of the detected influential user. This has opened up a wide range of opportunities to utilize the existing techniques to model and analyze the topic-level influence in online social networks. In this paper, we perform a comprehensive study of existing techniques used to infer the topic-based influential users in online social networks. We present a detailed review of these approaches in a taxonomy while highlighting the challenges and limitations associated with each technique. Moreover, we perform a detailed study of different evaluation techniques used in the literature to overcome the challenges that arise in evaluating topic-based IUD approaches. Furthermore, closely related research topics and open research questions in topic-based IUD are discussed to provide a deep understanding of the literature and future directions.