Kwan Hui Lim

Ngai Lam Ho, Kwan Hui Lim

Tour itinerary planning and recommendation are challenging problems for tourists visiting unfamiliar cities. Many tour recommendation algorithms only consider factors such as the location and popularity of Points of Interest (POIs) but their solutions may not align well with the user's own preferences and other location constraints. Additionally, these solutions do not take into consideration of the users' preference based on their past POIs selection. In this paper, we propose POIBERT, an algorithm for recommending personalized itineraries using the BERT language model on POIs. POIBERT builds upon the highly successful BERT language model with the novel adaptation of a language model to our itinerary recommendation task, alongside an iterative approach to generate consecutive POIs. Our recommendation algorithm is able to generate a sequence of POIs that optimizes time and users' preference in POI categories based on past trajectories from similar tourists. Our tour recommendation algorithm is modeled by adapting the itinerary recommendation problem to the sentence completion problem in natural language processing (NLP). We also innovate an iterative algorithm to generate travel itineraries that satisfies the time constraints which is most likely from past trajectories. Using a Flickr dataset of seven cities, experimental results show that our algorithm out-performs many sequence prediction algorithms based on measures in recall, precision and F1-scores.

Menglin Li, Kwan Hui Lim, Teng Guo et al.

POI-level geo-information of social posts is critical to many location-based applications and services. However, the multi-modality, complexity and diverse nature of social media data and their platforms limit the performance of inferring such fine-grained locations and their subsequent applications. To address this issue, we present a transformer-based general framework, which builds upon pre-trained language models and considers non-textual data, for social post geolocation at the POI level. To this end, inputs are categorized to handle different social data, and an optimal combination strategy is provided for feature representations. Moreover, a uniform representation of hierarchy is proposed to learn temporal information, and a concatenated version of encodings is employed to capture feature-wise positions better. Experimental results on various social datasets demonstrate that three variants of our proposed framework outperform multiple state-of-art baselines by a large margin in terms of accuracy and distance error metrics.

Xiang Qian Ong, Kwan Hui Lim

Understanding the skill sets and knowledge required for any career is of utmost importance, but it is increasingly challenging in today's dynamic world with rapid changes in terms of the tools and techniques used. Thus, it is especially important to be able to accurately identify the required skill sets for any job for better career insights and development. In this paper, we propose and develop the Skill Recommendation (SkillRec) system for recommending the relevant job skills required for a given job based on the job title. SkillRec collects and identify the skill set required for a job based on the job descriptions published by companies hiring for these roles. In addition to the data collection and pre-processing capabilities, SkillRec also utilises word/sentence embedding techniques for job title representation, alongside a feed-forward neural network for job skill recommendation based on the job title representation. Based on our preliminary experiments on a dataset of 6,000 job titles and descriptions, SkillRec shows a promising performance in terms of accuracy and F1-score.

Junhua Liu, Kwan Hui Lim, Kristin L. Wood et al.

Itinerary recommendation is a complex sequence prediction problem with numerous real-world applications. This task becomes even more challenging when considering the optimization of multiple user queuing times and crowd levels, as well as numerous involved parameters, such as attraction popularity, queuing time, walking time, and operating hours. Existing solutions typically focus on single-person perspectives and fail to address real-world issues resulting from natural crowd behavior, like the Selfish Routing problem. In this paper, we introduce the Strategic and Crowd-Aware Itinerary Recommendation (SCAIR) algorithm, which optimizes group utility in real-world settings. We model the route recommendation strategy as a Markov Decision Process and propose a State Encoding mechanism that enables real-time planning and allocation in linear time. We evaluate our algorithm against various competitive and realistic baselines using a theme park dataset, demonstrating that SCAIR outperforms these baselines in addressing the Selfish Routing problem across four theme parks.

Wenchuan Mu, Kwan Hui Lim

The automatic scoring of summaries is important as it guides the development of summarizers. Scoring is also complex, as it involves multiple aspects such as fluency, grammar, and even textual entailment with the source text. However, summary scoring has not been considered a machine learning task to study its accuracy and robustness. In this study, we place automatic scoring in the context of regression machine learning tasks and perform evasion attacks to explore its robustness. Attack systems predict a non-summary string from each input, and these non-summary strings achieve competitive scores with good summarizers on the most popular metrics: ROUGE, METEOR, and BERTScore. Attack systems also "outperform" state-of-the-art summarization methods on ROUGE-1 and ROUGE-L, and score the second-highest on METEOR. Furthermore, a BERTScore backdoor is observed: a simple trigger can score higher than any automatic summarization method. The evasion attacks in this work indicate the low robustness of current scoring systems at the system level. We hope that our highlighting of these proposed attacks will facilitate the development of summary scores.

Ngai Lam Ho, Roy Ka-Wei Lee, Kwan Hui Lim

An essential task for tourists having a pleasant holiday is to have a well-planned itinerary with relevant recommendations, especially when visiting unfamiliar cities. Many tour recommendation tools only take into account a limited number of factors, such as popular Points of Interest (POIs) and routing constraints. Consequently, the solutions they provide may not always align with the individual users of the system. We propose an iterative algorithm in this paper, namely: BTREC (BERT-based Trajectory Recommendation), that extends from the POIBERT embedding algorithm to recommend personalized itineraries on POIs using the BERT framework. Our BTREC algorithm incorporates users' demographic information alongside past POI visits into a modified BERT language model to recommend a personalized POI itinerary prediction given a pair of source and destination POIs. Our recommendation system can create a travel itinerary that maximizes POIs visited, while also taking into account user preferences for categories of POIs and time availability. Our recommendation algorithm is largely inspired by the problem of sentence completion in natural language processing (NLP). Using a dataset of eight cities of different sizes, our experimental results demonstrate that our proposed algorithm is stable and outperforms many other sequence prediction algorithms, measured by recall, precision, and F1-scores.

Ngai Lam Ho, Kwan Hui Lim

Tour itinerary recommendation involves planning a sequence of relevant Point-of-Interest (POIs), which combines challenges from the fields of both Operations Research (OR) and Recommendation Systems (RS). As an OR problem, there is the need to maximize a certain utility (e.g., popularity of POIs in the tour) while adhering to some constraints (e.g., maximum time for the tour). As a RS problem, it is heavily related to problem or filtering or ranking a subset of POIs that are relevant to a user and recommending it as part of an itinerary. In this paper, we explore the use of language models for the task of tour itinerary recommendation and planning. This task has the unique requirement of recommending personalized POIs relevant to users and planning these POIs as an itinerary that satisfies various constraints. We discuss some approaches in this area, such as using word embedding techniques like Word2Vec and GloVe for learning POI embeddings and transformer-based techniques like BERT for generating itineraries.

Ngai Lam Ho, Roy Ka-Wei Lee, Kwan Hui Lim

When traveling to an unfamiliar city for holidays, tourists often rely on guidebooks, travel websites, or recommendation systems to plan their daily itineraries and explore popular points of interest (POIs). However, these approaches may lack optimization in terms of time feasibility, localities, and user preferences. In this paper, we propose the SBTRec algorithm: a BERT-based Trajectory Recommendation with sentiment analysis, for recommending personalized sequences of POIs as itineraries. The key contributions of this work include analyzing users' check-ins and uploaded photos to understand the relationship between POI visits and distance. We introduce SBTRec, which encompasses sentiment analysis to improve recommendation accuracy by understanding users' preferences and satisfaction levels from reviews and comments about different POIs. Our proposed algorithms are evaluated against other sequence prediction methods using datasets from 8 cities. The results demonstrate that SBTRec achieves an average F1 score of 61.45%, outperforming baseline algorithms. The paper further discusses the flexibility of the SBTRec algorithm, its ability to adapt to different scenarios and cities without modification, and its potential for extension by incorporating additional information for more reliable predictions. Overall, SBTRec provides personalized and relevant POI recommendations, enhancing tourists' overall trip experiences. Future work includes fine-tuning personalized embeddings for users, with evaluation of users' comments on POIs,~to further enhance prediction accuracy.

Junhua Liu, Yong Keat Tan, Bin Fu et al.

In conversational AI systems, a critical challenge in training effective multi-turn intent classification models lies in the generation of large-scale, domain-specific, multilingual dialogue datasets. In this paper, we introduce Chain-of-Intent, a novel framework that integrates Hidden Markov Models (HMMs) with Large Language Models (LLMs) to generate intent-driven, context-aware dialogues through self-play. Our method first extracts domain-specific intent transition patterns from real-world e-commerce chat logs, which guide the modeling of turn-level dynamics and intent sequences. LLMs are then employed to parameterize the emission probabilities of HMMs, enabling the generation of natural, coherent utterances aligned with predicted intents and dialogue context. We also propose MINT-CL, a multi-task contrastive learning framework for multi-turn intent classification, which improves performance while reducing dependence on large-scale annotated datasets. Empirical results demonstrate that our approach outperforms competitive baselines in dialogue generation quality and classification accuracy, particularly in multilingual settings. To facilitate future research, we release MINT-E, a comprehensive, multilingual, intent-aware multi-turn dialogue corpus derived from the e-commerce domain\footnote{The reproduced source code and dataset are available at https://github.com/junhua/chain-of-intent.

Kunrong Li, Kwan Hui Lim

Next point-of-interest (POI) recommendation predicts a user's next destination from historical movements. Traditional models require intensive training, while LLMs offer flexible and generalizable zero-shot solutions but often generate generic or geographically irrelevant results due to missing trajectory and spatial context. To address these issues, we propose RALLM-POI, a framework that couples LLMs with retrieval-augmented generation and self-rectification. We first propose a Historical Trajectory Retriever (HTR) that retrieves relevant past trajectories to serve as contextual references, which are then reranked by a Geographical Distance Reranker (GDR) for prioritizing spatially relevant trajectories. Lastly, an Agentic LLM Rectifier (ALR) is designed to refine outputs through self-reflection. Without additional training, RALLM-POI achieves substantial accuracy gains across three real-world Foursquare datasets, outperforming both conventional and LLM-based baselines. Code is released at https://github.com/LKRcrocodile/RALLM-POI.

Menglin Li, Kwan Hui Lim

The Financial Relation Extraction (FinRE) task involves identifying the entities and their relation, given a piece of financial statement/text. To solve this FinRE problem, we propose a simple but effective strategy that improves the performance of pre-trained language models by augmenting them with Named Entity Recognition (NER) and Part-Of-Speech (POS), as well as different approaches to combine these information. Experiments on a financial relations dataset show promising results and highlights the benefits of incorporating NER and POS in existing models. Our dataset and codes are available at https://github.com/kwanhui/FinRelExtract.

Wenchuan Mu, Kwan Hui Lim

Academic writing should be concise as concise sentences better keep the readers' attention and convey meaning clearly. Writing concisely is challenging, for writers often struggle to revise their drafts. We introduce and formulate revising for concision as a natural language processing task at the sentence level. Revising for concision requires algorithms to use only necessary words to rewrite a sentence while preserving its meaning. The revised sentence should be evaluated according to its word choice, sentence structure, and organization. The revised sentence also needs to fulfil semantic retention and syntactic soundness. To aide these efforts, we curate and make available a benchmark parallel dataset that can depict revising for concision. The dataset contains 536 pairs of sentences before and after revising, and all pairs are collected from college writing centres. We also present and evaluate the approaches to this problem, which may assist researchers in this area.

Junhua Liu, Yang Jihao, Cheng Chang et al.

Proactive intent prediction is a critical capability in modern e-commerce chatbots, enabling "zero-query" recommendations by anticipating user needs from behavioral and contextual signals. However, existing industrial systems face two fundamental challenges: (1) the semantic gap between discrete user features and the semantic intents within the chatbot's Knowledge Base, and (2) the objective misalignment between general-purpose LLM outputs and task-specific ranking utilities. To address these issues, we propose RGAlign-Rec, a closed-loop alignment framework that integrates an LLM-based semantic reasoner with a Query-Enhanced (QE) ranking model. We also introduce Ranking-Guided Alignment (RGA), a multi-stage training paradigm that utilizes downstream ranking signals as feedback to refine the LLM's latent reasoning. Extensive experiments on a large-scale industrial dataset from Shopee demonstrate that RGAlign-Rec achieves a 0.12% gain in GAUC, leading to a significant 3.52% relative reduction in error rate, and a 0.56% improvement in Recall@3. Online A/B testing further validates the cumulative effectiveness of our framework: the Query-Enhanced model (QE-Rec) initially yields a 0.98% improvement in CTR, while the subsequent Ranking-Guided Alignment stage contributes an additional 0.13% gain. These results indicate that ranking-aware alignment effectively synchronizes semantic reasoning with ranking objectives, significantly enhancing both prediction accuracy and service quality in real-world proactive recommendation systems.

Junhua Liu, Roy Ka-Wei Lee, Kwan Hui Lim

Human decision-making in high-stakes domains often relies on expertise and heuristics, but is vulnerable to hard-to-detect cognitive biases that threaten fairness and long-term outcomes. This work presents a novel approach to enhancing complex decision-making workflows through the integration of hierarchical learning alongside various enhancements. Focusing on university admissions as a representative high-stakes domain, we propose BGM-HAN, an enhanced Byte-Pair Encoded, Gated Multi-head Hierarchical Attention Network, designed to effectively model semi-structured applicant data. BGM-HAN captures multi-level representations that are crucial for nuanced assessment, improving both interpretability and predictive performance. Experimental results on real admissions data demonstrate that our proposed model significantly outperforms both state-of-the-art baselines from traditional machine learning to large language models, offering a promising framework for augmenting decision-making in domains where structure, context, and fairness matter. Source code is available at: https://github.com/junhua/bgm-han.

Junhua Liu, Yung Chuen Ng, Kwan Hui Lim

There has been growing interest in utilizing occupational data mining and analysis. In today's job market, occupational data mining and analysis is growing in importance as it enables companies to predict employee turnover, model career trajectories, screen through resumes and perform other human resource tasks. A key requirement to facilitate these tasks is the need for an occupation-related dataset. However, most research use proprietary datasets or do not make their dataset publicly available, thus impeding development in this area. To solve this issue, we present the Industrial and Professional Occupation Dataset (IPOD), which comprises 192k job titles belonging to 56k LinkedIn users. In addition to making IPOD publicly available, we also: (i) manually annotate each job title with its associated level of seniority, domain of work and location; and (ii) provide embedding for job titles and discuss various use cases. This dataset is publicly available at https://github.com/junhua/ipod.

Junhua Liu, Yung Chuen Ng, Kristin L. Wood et al.

Occupational data mining and analysis is an important task in understanding today's industry and job market. Various machine learning techniques are proposed and gradually deployed to improve companies' operations for upstream tasks, such as employee churn prediction, career trajectory modelling and automated interview. Job titles analysis and embedding, as the fundamental building blocks, are crucial upstream tasks to address these occupational data mining and analysis problems. In this work, we present the Industrial and Professional Occupations Dataset (IPOD), which consists of over 190,000 job titles crawled from over 56,000 profiles from Linkedin. We also illustrate the usefulness of IPOD by addressing two challenging upstream tasks, including: (i) proposing Title2vec, a contextual job title vector representation using a bidirectional Language Model (biLM) approach; and (ii) addressing the important occupational Named Entity Recognition problem using Conditional Random Fields (CRF) and bidirectional Long Short-Term Memory with CRF (LSTM-CRF). Both CRF and LSTM-CRF outperform human and baselines in both exact-match accuracy and F1 scores. The dataset and pre-trained embeddings are available at https://www.github.com/junhua/ipod.

Junhua Liu, Yong Keat Tan, Bin Fu et al.

Multi-turn intent classification is notably challenging due to the complexity and evolving nature of conversational contexts. This paper introduces LARA, a Linguistic-Adaptive Retrieval-Augmentation framework to enhance accuracy in multi-turn classification tasks across six languages, accommodating a large number of intents in chatbot interactions. LARA combines a fine-tuned smaller model with a retrieval-augmented mechanism, integrated within the architecture of LLMs. The integration allows LARA to dynamically utilize past dialogues and relevant intents, thereby improving the understanding of the context. Furthermore, our adaptive retrieval techniques bolster the cross-lingual capabilities of LLMs without extensive retraining and fine-tuning. Comprehensive experiments demonstrate that LARA achieves state-of-the-art performance on multi-turn intent classification tasks, enhancing the average accuracy by 3.67\% from state-of-the-art single-turn intent classifiers.

Kunrong Li, Zhu Sun, Kwan Hui Lim

We propose HyMoERec, a novel sequential recommendation framework that addresses the limitations of uniform Position-wise Feed-Forward Networks in existing models. Current approaches treat all user interactions and items equally, overlooking the heterogeneity in user behavior patterns and diversity in item complexity. HyMoERec initially introduces a hybrid mixture-of-experts architecture that combines shared and specialized expert branches with an adaptive expert fusion mechanism for the sequential recommendation task. This design captures diverse reasoning for varied users and items while ensuring stable training. Experiments on MovieLens-1M and Beauty datasets demonstrate that HyMoERec consistently outperforms state-of-the-art baselines.

Junhua Liu, Justin Albrethsen, Lincoln Goh et al.

Resource allocation in tactical ad-hoc networks presents unique challenges due to their dynamic and multi-hop nature. Accurate prediction of future network connectivity is essential for effective resource allocation in such environments. In this paper, we introduce the Spatial-Temporal Graph Encoder-Decoder (STGED) framework for Tactical Communication Networks that leverages both spatial and temporal features of network states to learn latent tactical behaviors effectively. STGED hierarchically utilizes graph-based attention mechanism to spatially encode a series of communication network states, leverages a recurrent neural network to temporally encode the evolution of states, and a fully-connected feed-forward network to decode the connectivity in the future state. Through extensive experiments, we demonstrate that STGED consistently outperforms baseline models by large margins across different time-steps input, achieving an accuracy of up to 99.2\% for the future state prediction task of tactical communication networks.

Junhua Liu, Yong Keat Tan, Bin Fu et al.

Accurate multi-turn intent classification is essential for advancing conversational AI systems. However, challenges such as the scarcity of comprehensive datasets and the complexity of contextual dependencies across dialogue turns hinder progress. This paper presents two novel approaches leveraging Large Language Models (LLMs) to enhance scalability and reduce latency in production dialogue systems. First, we introduce Symbol Tuning, which simplifies intent labels to reduce task complexity and improve performance in multi-turn dialogues. Second, we propose C-LARA (Consistency-aware, Linguistics Adaptive Retrieval Augmentation), a framework that employs LLMs for data augmentation and pseudo-labeling to generate synthetic multi-turn dialogues. These enriched datasets are used to fine-tune a small, efficient model suitable for deployment. Experiments conducted on multilingual dialogue datasets demonstrate significant improvements in classification accuracy and resource efficiency. Our methods enhance multi-turn intent classification accuracy by 5.09%, reduce annotation costs by 40%, and enable scalable deployment in low-resource multilingual industrial systems, highlighting their practicality and impact.

Wenchuan Mu, Kwan Hui Lim

In today's data and information-rich world, summarization techniques are essential in harnessing vast text to extract key information and enhance decision-making and efficiency. In particular, topic-focused summarization is important due to its ability to tailor content to specific aspects of an extended text. However, this usually requires extensive labelled datasets and considerable computational power. This study introduces a novel method, Augmented-Query Summarization (AQS), for topic-focused summarization without the need for extensive labelled datasets, leveraging query augmentation and hierarchical clustering. This approach facilitates the transferability of machine learning models to the task of summarization, circumventing the need for topic-specific training. Through real-world tests, our method demonstrates the ability to generate relevant and accurate summaries, showing its potential as a cost-effective solution in data-rich environments. This innovation paves the way for broader application and accessibility in the field of topic-focused summarization technology, offering a scalable, efficient method for personalized content extraction.

Wenchuan Mu, Kwan Hui Lim

In safety-critical deep learning applications, robustness measures the ability of neural models that handle imperceptible perturbations in input data, which may lead to potential safety hazards. Existing pre-deployment robustness assessment methods typically suffer from significant trade-offs between computational cost and measurement precision, limiting their practical utility. To address these limitations, this paper conducts a comprehensive comparative analysis of existing robustness definitions and associated assessment methodologies. We propose tower robustness to evaluate robustness, which is a novel, practical metric based on hypothesis testing to quantitatively evaluate probabilistic robustness, enabling more rigorous and efficient pre-deployment assessments. Our extensive comparative evaluation illustrates the advantages and applicability of our proposed approach, thereby advancing the systematic understanding and enhancement of model robustness in safety-critical deep learning applications.

Junhua Liu, Fanfan Lin, Xinze Li et al.

LLM-based autonomous agents have recently shown strong capabilities in solving complex industrial design tasks. However, in domains aiming for carbon neutrality and high-performance renewable energy systems, current AI-assisted design automation methods face critical challenges in explainability, scalability, and practical usability. To address these limitations, we introduce PHIA (Physics-Informed Autonomous Agent), an LLM-driven system that automates modulation design for power converters in Power Electronics Systems with minimal human intervention. In contrast to traditional pipeline-based methods, PHIA incorporates an LLM-based planning module that interactively acquires and verifies design requirements via a user-friendly chat interface. This planner collaborates with physics-informed simulation and optimization components to autonomously generate and iteratively refine modulation designs. The interactive interface also supports interpretability by providing textual explanations and visual outputs throughout the design process. Experimental results show that PHIA reduces standard mean absolute error by 63.2% compared to the second-best benchmark and accelerates the overall design process by over 33 times. A user study involving 20 domain experts further confirms PHIA's superior design efficiency and usability, highlighting its potential to transform industrial design workflows in power electronics.

Wenchuan Mu, Kwan Hui Lim

In deep learning applications, robustness measures the ability of neural models that handle slight changes in input data, which could lead to potential safety hazards, especially in safety-critical applications. Pre-deployment assessment of model robustness is essential, but existing methods often suffer from either high costs or imprecise results. To enhance safety in real-world scenarios, metrics that effectively capture the model's robustness are needed. To address this issue, we compare the rigour and usage conditions of various assessment methods based on different definitions. Then, we propose a straightforward and practical metric utilizing hypothesis testing for probabilistic robustness and have integrated it into the TorchAttacks library. Through a comparative analysis of diverse robustness assessment methods, our approach contributes to a deeper understanding of model robustness in safety-critical applications.

Junhua Liu, Kwan Hui Lim, Roy Ka-Wei Lee

How objective and unbiased are we while making decisions? This work investigates cognitive bias identification in high-stake decision making process by human experts, questioning its effectiveness in real-world settings, such as candidates assessments for university admission. We begin with a statistical analysis assessing correlations among different decision points among in the current process, which discovers discrepancies that imply cognitive bias and inconsistency in decisions. This motivates our exploration of bias-aware AI-augmented workflow that surpass human judgment. We propose BGM-HAN, an enhanced Hierarchical Attention Network with Byte-Pair Encoding, Gated Residual Connections and Multi-Head Attention. Using it as a backbone model, we further propose a Shortlist-Analyse-Recommend (SAR) agentic workflow, which simulate real-world decision-making. In our experiments, both the proposed model and the agentic workflow significantly improves on both human judgment and alternative models, validated with real-world data.

Henry Lim, Kwan Hui Lim

Instruction-tuned large language models (IT-LLMs) exhibit strong zero-shot reasoning, yet their ability to execute simple, self-contained instructions remains underexplored, despite this being foundational to complex instruction-following. We evaluate 20 IT-LLMs on modified MMLU and MMLU-Pro benchmarks, by systematically varying the format of option labels (alphabetic, numeric, Roman) while keeping their meaning identical under four paradigms, namely: (1) With explicit instructions, label changes cause large performance shifts (e.g., -30.45\% for Roman vs. numeric), revealing instruction-format bias. (2) Without instructions, performance drops further (up to -10.84\%) and label sensitivity intensifies, underscoring the role of explicit guidance. (3) When option contents are removed, models fail random-choice baselines except with numeric labels, suggesting weak adherence to atomic directives. (4) Three-shot exemplars yield no significant gains in robustness or fidelity, and generation analyses show persistent label errors, especially for non-numeric formats. Across model sizes, larger LLMs achieve higher accuracy but remain inconsistent in instruction adherence. These results expose the insufficiencies of current instruction-tuning paradigms and highlight the need for evaluation methods and training strategies that explicitly target atomic instruction-following.

Wenchuan Mu, Menglin Li, Kwan Hui Lim

Social media platforms such as Twitter and Facebook have become deeply embedded in our everyday life, offering a dynamic stream of localized news and personal experiences. The ubiquity of these platforms position them as valuable resources for identifying estate-related issues, especially in the context of growing urban populations. In this work, we present a language model-based system for the detection and classification of estate-related events from social media content. Our system employs a hierarchical classification framework to first filter relevant posts and then categorize them into actionable estate-related topics. Additionally, for posts lacking explicit geotags, we apply a transformer-based geolocation module to infer posting locations at the point-of-interest level. This integrated approach supports timely, data-driven insights for urban management, operational response and situational awareness.

Junhua Liu, Roy Ka-Wei Lee, Kwan Hui Lim

Fairness in both Machine Learning (ML) predictions and human decision-making is essential, yet both are susceptible to different forms of bias, such as algorithmic and data-driven in ML, and cognitive or subjective in humans. In this study, we examine fairness using a real-world university admissions dataset comprising 870 applicant profiles, leveraging three ML models: XGB, Bi-LSTM, and KNN, alongside BERT embeddings for textual features. To evaluate individual fairness, we introduce a consistency metric that quantifies agreement in decisions among ML models and human experts with diverse backgrounds. Our analysis reveals that ML models surpass human evaluators in fairness consistency by margins ranging from 14.08\% to 18.79\%. Our findings highlight the potential of using ML to enhance fairness in admissions while maintaining high accuracy, advocating a hybrid approach combining human judgement and ML models.

Menglin Li, Kwan Hui Lim

To address the challenges of scarcity in geotagged data for social user geolocation, we propose FewUser, a novel framework for Few-shot social User geolocation. We incorporate a contrastive learning strategy between users and locations to improve geolocation performance with no or limited training data. FewUser features a user representation module that harnesses a pre-trained language model (PLM) and a user encoder to process and fuse diverse social media inputs effectively. To bridge the gap between PLM's knowledge and geographical data, we introduce a geographical prompting module with hard, soft, and semi-soft prompts, to enhance the encoding of location information. Contrastive learning is implemented through a contrastive loss and a matching loss, complemented by a hard negative mining strategy to refine the learning process. We construct two datasets TwiU and FliU, containing richer metadata than existing benchmarks, to evaluate FewUser and the extensive experiments demonstrate that FewUser significantly outperforms state-of-the-art methods in both zero-shot and various few-shot settings, achieving absolute improvements of 26.95\% and \textbf{41.62\%} on TwiU and FliU, respectively, with only one training sample per class. We further conduct a comprehensive analysis to investigate the impact of user representation on geolocation performance and the effectiveness of FewUser's components, offering valuable insights for future research in this area.

Trisha Singhal, Junhua Liu, Lucienne T. M. Blessing et al.

The scientific world is changing at a rapid pace, with new technology being developed and new trends being set at an increasing frequency. This paper presents a framework for conducting scientific analyses of academic publications, which is crucial to monitor research trends and identify potential innovations. This framework adopts and combines various techniques of Natural Language Processing, such as word embedding and topic modelling. Word embedding is used to capture semantic meanings of domain-specific words. We propose two novel scientific publication embedding, i.e., PUB-G and PUB-W, which are capable of learning semantic meanings of general as well as domain-specific words in various research fields. Thereafter, topic modelling is used to identify clusters of research topics within these larger research fields. We curated a publication dataset consisting of two conferences and two journals from 1995 to 2020 from two research domains. Experimental results show that our PUB-G and PUB-W embeddings are superior in comparison to other baseline embeddings by a margin of ~0.18-1.03 based on topic coherence.

Yasmeen George, Shanika Karunasekera, Aaron Harwood et al.

A key challenge in mining social media data streams is to identify events which are actively discussed by a group of people in a specific local or global area. Such events are useful for early warning for accident, protest, election or breaking news. However, neither the list of events nor the resolution of both event time and space is fixed or known beforehand. In this work, we propose an online spatio-temporal event detection system using social media that is able to detect events at different time and space resolutions. First, to address the challenge related to the unknown spatial resolution of events, a quad-tree method is exploited in order to split the geographical space into multiscale regions based on the density of social media data. Then, a statistical unsupervised approach is performed that involves Poisson distribution and a smoothing method for highlighting regions with unexpected density of social posts. Further, event duration is precisely estimated by merging events happening in the same region at consecutive time intervals. A post processing stage is introduced to filter out events that are spam, fake or wrong. Finally, we incorporate simple semantics by using social media entities to assess the integrity, and accuracy of detected events. The proposed method is evaluated using different social media datasets: Twitter and Flickr for different cities: Melbourne, London, Paris and New York. To verify the effectiveness of the proposed method, we compare our results with two baseline algorithms based on fixed split of geographical space and clustering method. For performance evaluation, we manually compute recall and precision. We also propose a new quality measure named strength index, which automatically measures how accurate the reported event is.

Prashant Solanki, Kwan Hui Lim, Aaron Harwood

With the prevalence of online social networking sites (OSNs) and mobile devices, people are increasingly reliant on a variety of OSNs for keeping in touch with family and friends, and using it as a source of information. For example, a user might utilise multiple OSNs for different purposes, such as using Flickr to share holiday pictures with family and friends, and Twitter to post short messages about their thoughts. Identifying the same user across multiple OSNs is an important task as this allows us to understand the usage patterns of users among different OSNs, make recommendations when a user registers for a new OSN, and various other useful applications. To address this problem, we proposed an algorithm based on the multilayer perceptron using various types of features, namely: (i) user profile, such as name, location, description; (ii) temporal distribution of user generated content; and (iii) embedding based on user name, real name and description. Using a Twitter and Flickr dataset of users and their posting activities, we perform an empirical study on how these features affect the performance of user identification across the two OSNs and discuss our main findings based on the different features.

Trisha Singhal, Junhua Liu, Lucienne T. M. Blessing et al.

The advent of social media platforms has been a catalyst for the development of digital photography that engendered a boom in vision applications. With this motivation, we introduce a large-scale dataset termed 'Photozilla', which includes over 990k images belonging to 10 different photographic styles. The dataset is then used to train 3 classification models to automatically classify the images into the relevant style which resulted in an accuracy of ~96%. With the rapid evolution of digital photography, we have seen new types of photography styles emerging at an exponential rate. On that account, we present a novel Siamese-based network that uses the trained classification models as the base architecture to adapt and classify unseen styles with only 25 training samples. We report an accuracy of over 68% for identifying 10 other distinct types of photography styles. This dataset can be found at https://trisha025.github.io/Photozilla/

Ngai Lam Ho, Kwan Hui Lim

Tour itinerary planning and recommendation are challenging tasks for tourists in unfamiliar countries. Many tour recommenders only consider broad POI categories and do not align well with users' preferences and other locational constraints. We propose an algorithm to recommend personalized tours using POI-embedding methods, which provides a finer representation of POI types. Our recommendation algorithm will generate a sequence of POIs that optimizes time and locational constraints, as well as user's preferences based on past trajectories from similar tourists. Our tour recommendation algorithm is modelled as a word embedding model in natural language processing, coupled with an iterative algorithm for generating itineraries that satisfies time constraints. Using a Flickr dataset of 4 cities, preliminary experimental results show that our algorithm is able to recommend a relevant and accurate itinerary, based on measures of recall, precision and F1-scores.

Jolin Shaynn-Ly Kwan, Kwan Hui Lim

The COVID-19 pandemic has created widespread health and economical impacts, affecting millions around the world. To better understand these impacts, we present the TweetCOVID system that offers the capability to understand the public reactions to the COVID-19 pandemic in terms of their sentiments, emotions, topics of interest and controversial discussions, over a range of time periods and locations, using public tweets. We also present three example use cases that illustrates the usefulness of our proposed TweetCOVID system.

Jerome Heng, Junhua Liu, Kwan Hui Lim

An important aspect of urban planning is understanding crowd levels at various locations, which typically require the use of physical sensors. Such sensors are potentially costly and time consuming to implement on a large scale. To address this issue, we utilize publicly available social media datasets and use them as the basis for two urban sensing problems, namely event detection and crowd level prediction. One main contribution of this work is our collected dataset from Twitter and Flickr, alongside ground truth events. We demonstrate the usefulness of this dataset with two preliminary supervised learning approaches: firstly, a series of neural network models to determine if a social media post is related to an event and secondly a regression model using social media post counts to predict actual crowd levels. We discuss preliminary results from these tasks and highlight some challenges.

Yan Zhang, Ruidan He, Zuozhu Liu et al.

BERT is inefficient for sentence-pair tasks such as clustering or semantic search as it needs to evaluate combinatorially many sentence pairs which is very time-consuming. Sentence BERT (SBERT) attempted to solve this challenge by learning semantically meaningful representations of single sentences, such that similarity comparison can be easily accessed. However, SBERT is trained on corpus with high-quality labeled sentence pairs, which limits its application to tasks where labeled data is extremely scarce. In this paper, we propose a lightweight extension on top of BERT and a novel self-supervised learning objective based on mutual information maximization strategies to derive meaningful sentence embeddings in an unsupervised manner. Unlike SBERT, our method is not restricted by the availability of labeled data, such that it can be applied on different domain-specific corpus. Experimental results show that the proposed method significantly outperforms other unsupervised sentence embedding baselines on common semantic textual similarity (STS) tasks and downstream supervised tasks. It also outperforms SBERT in a setting where in-domain labeled data is not available, and achieves performance competitive with supervised methods on various tasks.

Junhua Liu, Trisha Singhal, Lucienne T. M. Blessing et al.

Since the start of COVID-19, several relevant corpora from various sources are presented in the literature that contain millions of data points. While these corpora are valuable in supporting many analyses on this specific pandemic, researchers require additional benchmark corpora that contain other epidemics to facilitate cross-epidemic pattern recognition and trend analysis tasks. During our other efforts on COVID-19 related work, we discover very little disease related corpora in the literature that are sizable and rich enough to support such cross-epidemic analysis tasks. In this paper, we present EPIC30M, a large-scale epidemic corpus that contains 30 millions micro-blog posts, i.e., tweets crawled from Twitter, from year 2006 to 2020. EPIC30M contains a subset of 26.2 millions tweets related to three general diseases, namely Ebola, Cholera and Swine Flu, and another subset of 4.7 millions tweets of six global epidemic outbreaks, including 2009 H1N1 Swine Flu, 2010 Haiti Cholera, 2012 Middle-East Respiratory Syndrome (MERS), 2013 West African Ebola, 2016 Yemen Cholera and 2018 Kivu Ebola. Furthermore, we explore and discuss the properties of the corpus with statistics of key terms and hashtags and trends analysis for each subset. Finally, we demonstrate the value and impact that EPIC30M could create through a discussion of multiple use cases of cross-epidemic research topics that attract growing interest in recent years. These use cases span multiple research areas, such as epidemiological modeling, pattern recognition, natural language understanding and economical modeling.

Junhua Liu, Trisha Singhal, Lucienne T. M. Blessing et al.

Classification of crisis events, such as natural disasters, terrorist attacks and pandemics, is a crucial task to create early signals and inform relevant parties for spontaneous actions to reduce overall damage. Despite crisis such as natural disasters can be predicted by professional institutions, certain events are first signaled by civilians, such as the recent COVID-19 pandemics. Social media platforms such as Twitter often exposes firsthand signals on such crises through high volume information exchange over half a billion tweets posted daily. Prior works proposed various crisis embeddings and classification using conventional Machine Learning and Neural Network models. However, none of the works perform crisis embedding and classification using state of the art attention-based deep neural networks models, such as Transformers and document-level contextual embeddings. This work proposes CrisisBERT, an end-to-end transformer-based model for two crisis classification tasks, namely crisis detection and crisis recognition, which shows promising results across accuracy and f1 scores. The proposed model also demonstrates superior robustness over benchmark, as it shows marginal performance compromise while extending from 6 to 36 events with only 51.4% additional data points. We also proposed Crisis2Vec, an attention-based, document-level contextual embedding architecture for crisis embedding, which achieve better performance than conventional crisis embedding methods such as Word2Vec and GloVe. To the best of our knowledge, our works are first to propose using transformer-based crisis classification and document-level contextual crisis embedding in the literature.

Junhua Liu, Kristin L. Wood, Kwan Hui Lim

There is a rapidly growing demand for itinerary planning in tourism but this task remains complex and difficult, especially when considering the need to optimize for queuing time and crowd levels for multiple users. This difficulty is further complicated by the large amount of parameters involved, i.e., attraction popularity, queuing time, walking time, operating hours, etc. Many recent works propose solutions based on the single-person perspective, but otherwise do not address real-world problems resulting from natural crowd behavior, such as the Selfish Routing problem, which describes the consequence of ineffective network and sub-optimal social outcome by leaving agents to decide freely. In this work, we propose the Strategic and Crowd-Aware Itinerary Recommendation (SCAIR) algorithm which optimizes social welfare in real-world situations. We formulate the strategy of route recommendation as Markov chains which enables our simulations to be carried out in poly-time. We then evaluate our proposed algorithm against various competitive and realistic baselines using a theme park dataset. Our simulation results highlight the existence of the Selfish Routing problem and show that SCAIR outperforms the baselines in handling this issue.