A comprehensive review on resolving ambiguities in natural language processing

Yadav, Apurwa; Patel, Aarshil; Shah, Manan

doi:10.1016/j.aiopen.2021.05.001

Cited by 22 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, requirements written in natural language are often ambiguous. The ambiguity may lead to software failures and a waste of time (Yadav et al 2021). Using formal requirement specification languages can rarely deal with ambiguity well, often resulting in misunderstandings being documented in the 10 unambiguous requirements (Kamsties et al 2001).…”

Section: Ambiguity In Requirements Engineeringmentioning

confidence: 99%

Unpacking Ambiguity in Building Requirements to Support Automated Compliance Checking

Zhang

Nisbet

2023

J. Manage. Eng.

View full text Add to dashboard Cite

In the architectural, engineering and construction (AEC) industry, manual compliance checking is labour-intensive, time-consuming, expensive and error-prone. Automated compliance checking (ACC) has been extensively studied in the past 50 years to improve the productivity and accuracy of the compliance checking process. While numerous ACC systems have been proposed, these systems can only deal with requirements that include quantitative metrics or specified properties. This leaves the remaining 53% of the building requirements to be checked manually, mainly due to the ambiguity embedded in them. In the literature, little is known about the ambiguity of building requirements, which impedes the accurate interpretation and automated checking of these requirements. This research thus aims to address this issue and establish a taxonomy of ambiguity. Building requirements in Health Building Notes (HBN) are analysed using an inductive approach. The results show that some ambiguous clauses in building requirements reflect regulators' intention, while others are unintentional, resulting from the use of language, tacit knowledge and ACC-specific reasons. This research is valuable for compliance-checking researchers and practitioners by unpacking ambiguity in building requirements, which lays a solid foundation to address ambiguity appropriately.

show abstract

Section: Ambiguity In Requirements Engineeringmentioning

confidence: 99%

Unpacking Ambiguity in Building Requirements to Support Automated Compliance Checking

Zhang

Nisbet

2023

J. Manage. Eng.

View full text Add to dashboard Cite

show abstract

“…This issue of ambiguity in text-image translation can be solved by establishing an intermediary medium for delivering user expectations. 8 Language-based models can be applied to tackle specific design aspects and not the holistic complexity of the design activity. Generative deep learning models, such as StyleGAN, can be used for different design tasks, and a language model (i.e., CLIP) for certain tasks within the design workflow.…”

Section: Shermeen Yousifmentioning

confidence: 99%

Using Language-Based and Generative Deep Learning Models for Encoding Design Intentions and Modifying Architectural Design

Yousif

2022

Empower

View full text Add to dashboard Cite

Artificial intelligence models are moving design exploration beyond the deterministic rule-based parametric systems by offering new possibilities and expanding the design space, which has become more flexible and adaptive to change. Yet, the fact that AI models are independently learning on their own, raises issues with designers’ control over the process. More recently, models that bridge natural language processing and computer-vision such as Contrastive Language-Image Pre-Training (CLIP) have been integrated into generative deep learning models such as StyleGAN, combining the generative and classification functionalities. This way, to some degree, a certain level of designer’s agency can be attained when using text prompts to modify the generative process, which was the motivation of this work. We investigate here the issue of prototyping a new design system with employing language-based models and deep learning models into an expanded design space towards informing design revision and modification. Our methodology involves experimenting with the targeted deep learning models, prototyping a new framework with language-based models are integrated into the generative process, and testing the prototype by applying the proposed system to a design case. As a result of experimentation, the generative model was modified using a set of text-prompts that describe the intended design alteration. Overall, the results show successful approaches to guiding the generative process and informing design revision, and offer insights into associated potentials and limitations, as discussed in the paper.

show abstract

“…They can be trained to determine that two texts are semantically identical even when the wording is significantly different. [3]. A model of this kind was developed [4]- [6] and utilized to group customer input into subjects, resulting in a more compelling customer feedback monitoring process.…”

Section: Introductionmentioning

confidence: 99%

Cluster Analysis of Japanese Whiskey Product Review Using K-Means Clustering

Witarsyah,

Akbar,

Praditha

et al. 2024

JOIV : Int. J. Inform. Visualization

View full text Add to dashboard Cite

Since 2008, the Japanese whiskey business has grown steadily. Overall, the whiskey market (at factory price) is expected to reach $2.95 billion in 2019, accounting for 8.6 percent of the entire alcoholic beverage industry. The rise in popularity of Japanese whiskey is associated with the country's growing international reputation. Founded 1985 as an independent bottler, Master of Malt was the first company to service clients who ordered single malt whiskey through the mail-order system. Master of Malt's omnichannel approach encompasses all channels available to the company. Known as their 'omnichannel,' this refers to the organization's capability to provide speed and precision from any place at any time. As their brand has grown over the years, they have used various marketing strategies, including a website redesign and rebuild that involved the creation of all relevant content and designing and constructing landing pages for their website. Following a clustering technique, we discovered that the data is being divided into four distinct groups and that these clusters may serve as a recommender system based on the occurrence of terms in each of the categories. Our summarizing component combined phrases related to the exact subtopics and provided users with a concise summary and sentimental information about the group of phrases.

show abstract

A comprehensive review on resolving ambiguities in natural language processing

Cited by 22 publications

References 10 publications

Unpacking Ambiguity in Building Requirements to Support Automated Compliance Checking

Unpacking Ambiguity in Building Requirements to Support Automated Compliance Checking

Using Language-Based and Generative Deep Learning Models for Encoding Design Intentions and Modifying Architectural Design

Cluster Analysis of Japanese Whiskey Product Review Using K-Means Clustering

Contact Info

Product

Resources

About