Domain knowledge as attention fixer in Large Language Models Y. Han, M. Moghaddam
Aspect-based sentiment analysis (ABSA) enables a systematic identification of user opinions on particular aspects, thus enhancing the idea creation process in the initial stages of product/service design. Attention-based large language models (LLMs) like T5 and GPT4 have proven powerful in ABSA tasks. Yet, several key limitations remain, both regarding the ABSA task and the capabilities of attention-based models. First, existing research mainly focuses on relatively simpler ABSA tasks such as aspect--based sentiment analysis, while the task of extracting aspect, opinion, and sentiment in a unified model remains largely unaddressed. Second, current ABSA tasks overlook implicit opinions and sentiments. Third, most attention-based LLMs like BERT use position encoding in a linear projected manner or through split-position relations in word distance schemes, which could lead to relation biases during the training process. This paper addresses these gaps by (1) creating a new annotated dataset with five types of labels, including aspect, category, opinion, sentiment, and implicit indicator (ACOSI), (2) developing a unified model capable of extracting all five types of labels simultaneously in a generative manner, and (3) designing a new position encoding method in the attention-based model. (4) introduced a new benchmark based on Rouge scoure that incorporate with design domain knowledge inside. The numerical experiments conducted on a manually labeled dataset scraped from three major e-Commerce retail stores for apparel and footwear products demonstrate the performance, scalability, and potential of the framework developed. The article concludes with recommendations for future research on automated need finding and sentiment analysis for user-centered design.
Aspect Guided Abstractive Summarization for Safety Concern Information Extration J. Shi, Y. Han
Aspect-guided summarization focuses on extracting information relevant to specific aspects from multiple documents. Unlike generic summarization and entity-related summarization, aspect-guided summarization differs in terms of granularity. In this paper, we introduce the SafetySum dataset and safety concerns-guided double encoder transformer (SDT) model, which specifically caters to aspect-guided summarization in the domain of safety engineering problems. We thoroughly analyze existing methods and demonstrate that both entity-centric summarization and controllable summarization techniques fall short of effectively addressing the requirements of aspect-guided summarization. To address this challenge, we experiment with previous approaches that can be adapted to this task and prove our SDT is state-of-the-art for this task. Our analysis underscores the difficulty of this task and highlights the need for innovative solutions.
Attribute-Sentiment-Guided Summarization of User Opinions from Online Reviews. Y. Han, G. Nanda, M. Moghaddam
Eliciting informative user opinions from online reviews is a key success factor for innovative product design and development. The unstructured, noisy, and verbose nature of user reviews, however, often complicate large-scale need finding in a format useful for designers without losing important information. Recent advances in abstractive text summarization have created the opportunity to systematically generate opinion summaries from online reviews to inform the early stages of product design and development. However, two knowledge gaps hinder the applicability of opinion summarization methods in practice. First, there is a lack of formal mechanisms to guide the generative process with respect to different categories of product attributes and user sentiments. Second, the annotated training datasets needed for supervised training of abstractive summarization models are often difficult and costly to create. This article addresses these gaps by (1) devising an efficient computational framework for abstractive opinion summarization guided by specific product attributes and sentiment polarities, and (2) automatically generating a synthetic training dataset that captures various degrees of granularity and polarity. A hierarchical multi-instance attribute-sentiment inference model is developed for assembling a high-quality synthetic dataset, which is utilized to fine-tune a pretrained language model for abstractive summary generation. Numerical experiments conducted on a large dataset scraped from three major e-Commerce retail stores for apparel and footwear products indicate the performance, feasibility, and potentials of the developed framework. Several directions are provided for future exploration in the area of automated opinion summarization for user-centered design.
Eliciting Attribute-Level User Needs From Online Reviews With Deep Language Models and Information Extraction. Yi Han, M.Moghaddam
Eliciting user needs for individual components and features of a product or a service on a large scale is a key requirement for innovative design. Synthesizing data as an initial discovery phase of a design process is usually accomplished with a small number of participants, employing qualitative research methods such as observations, focus groups, and interviews. This leaves an entire swath of pertinent user behavior, preferences, and opinions not captured. Sentiment analysis is a key enabler for large-scale need finding from online user reviews generated on a regular basis. A major limitation of current sentiment analysis approaches used in design sciences, however, is the need for laborious labeling and annotation of large review datasets for training, which in turn hinders their scalability and transferability across different domains. This article proposes an efficient and scalable methodology for automated and large-scale elicitation of attribute-level user needs. The methodology builds on the state-of-the-art pretrained deep language model, BERT (Bidirectional Encoder Representations from Transformers), with new convolutional net and named entity recognition (NER) layers for extracting attribute, description, and sentiment words from online user review corpora. The machine translation algorithm BLEU (BiLingual Evaluation Understudy) is utilized to extract need expressions in the form of predefined part-of-speech combinations (e.g., adjective–noun, verb–noun). Numerical experiments are conducted on a large dataset scraped from a major e-commerce retail store for apparel and footwear to demonstrate the performance, feasibility, and potentials of the developed methodology.
Analysis of sentiment expressions for user-centered design. Y. Han, M. Moghaddam
Devising intelligent systems capable of identifying the idiosyncratic needs of users at scale and translating them into attribute-level design feedback and recommendations is a key prerequisite for successful user-centered design processes. Recent studies show that 49% of design firms lack systems and tools for monitoring external platforms, and only 8% have adopted digital, data-driven approaches for new product development despite acknowledging them as a high priority. The state-of-the-art attribute-level sentiment analysis approaches based on deep learning have achieved promising results; however, these methods pose strict preconditions, require manually labeled data for training and pre-defined attributes by experts, and only classify sentiments intro predefined categories which have limited implications for designers. This article develops a rule-based methodology for extracting and analyzing the sentiment expressions of users on a large scale, from myriad reviews available on social media and e-commerce platforms. The methodology further advances current unsupervised attribute-level sentiment analysis approaches by enabling efficient identification and mapping of sentiment expressions of individual users onto their respective attributes. Experiments on a large dataset scraped from a major e-commerce retail store for apparel and indicate 74.3%–93.8% precision in extracting attribute-level sentiment expressions of users and demonstrate the feasibility and potentials of the developed methodology for large-scale need finding from user reviews.
Conference Papers
Shoes-ACOSI: A Dataset for Aspect-Based Sentiment Analysis with Implicit Opinion Extraction Joseph J Peper, Wenzhao Qiu, Ryan Bruggeman, Yi Han, Estefania Ciliotta Chehade, Lu Wang EMNLP 2024 Hyatt Regency Miami Hotel in Miami, Florida November 12–16, 2024
A Priori: Design Knowledge in AI R. Bruggeman, E.C. Chehade, Y. Han, P. Ciuccarelli The 12th International Conference on Design and Semantics of Form and Movement HongKong Jul.5-7 2023
A Design Knowledge Guided Position Encoding Methodology for Implicit Need Identification From User Reviews Y. Han, M. Moghaddam IDETC-CIE 2023 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference Boston Park Plaza, Boston MA Aug.20-23 2023
A Priori: Design Knowledge in AI R. Bruggeman, E.C. Chehade, Y. Han, P. Ciuccarelli The 12th International Conference on Design and Semantics of Form and Movement HongKong Jul.5-7 2023
Extracting latent needs from online reviews through deep learning based language model Y. Han, R. Bruggeman, J. Peper, E.C. Chehade, T. Marion, P. Ciuccarelli, M. Moghaddam 24th International Conference on Engineering Design Bordeaux, France Jul. 24-28 2023
Aspect-Sentiment-Guided Opinion Summarization for User Need Elicitation From Online Reviews Y. Han, M. Moghaddam IDETC-CIE 2022 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference St. Louis Union Station Hotel, St. Louis, Missouri Aug. 14–17, 2022
han.yi1@northeastern.edu
