China
Africa
Explore chapters and articles related to this topic
Semantic Web Technologies
Published in Archana Patel, Narayan C. Debnath, Bharat Bhushan, Semantic Web Technologies, 2023
Esingbemi P. Ebietomere, Godspower O. Ekuobase
As earlier implied, natural language processing (NLP) and semantic web have different but complementary goals. Thus, while NLP has found application in the semantic web, the semantic web has also helped revolutionize NLP. Natural language understanding (NLU) also referred to as semantic analysis [34] aims to represent the semantics (meaning) of a given text toward enabling machine readability and comprehension [25,34]. NLU has become one of the approaches to the semantic web, particularly in the area of knowledge management. It has found its role in automating knowledge acquisition tasks [57] thereby solving the problem of knowledge acquisition bottleneck – effort and time [56] that bedeviled explicit knowledge description of ontology at the early stage [58,59]. Also, the impact of NLU is visible in semantic web search, information visualization, and connecting text to linked open data [34,57]. Some of the SWTs that support NLU are highlighted under the section dedicated to semantic web tools.
AI-Informed Analytics Cycle: Reinforcing Concepts
Published in Jay Liebowitz, Data Analytics and AI, 2020
Rosina O. Weber, Maureen P. Kinkela
Although the main task seems to be predictive analytics, the analyst has to recognize that when data is unstructured, there is another large body of AI tasks and methods that require examination. When the data is textual, the methods required are from natural language understanding (NLU). In NLU, the complex tasks and methods are different from those described for an AI-informed analytics process when the data is structured. They are different because the real complex task is understanding, which follows a very different structure. In NLU, the primary concern is to learn a language model. The language model is what guides the decisions on how to execute multiple tasks such as search, or comparisons between words, segments, or documents. This chapter is limited in scope and does not cover textual methods, which would require another entire chapter. We briefly mention some of these methods in this illustration for the reader to appreciate the extent as well as the limitations of AI methods at the time this chapter was published.
Artificial Intelligence for Customer Experience and Care
Published in Mazin Gilbert, Artificial Intelligence for Autonomous Networks, 2018
Customers’ expectations can range from exceeded to not met. These two extreme ends of customer expectations might generate emotions that reflect sadness, happiness, and anger. The AI-trained systems can use the tone and sentiment analysis from user responses to understand the tone and sentiment of the customer. One such service is the Watson Tone Analyzer service, which is a pretrained service that can detect the tone in a spoken and written language. The tone can be positive, analytical, anger, confident, fear, or tentative. Another service like the IBM Watson Natural Language Understanding can detect the sentiment in a written text or a spoken language to be either positive, negative, or neutral. The combination of the sentiment and the tone can be augmented in a machine learning model to understand the customer’s mood and personality. As explained in the article on tone analyzer [7] “The Tone Analyzer was developed with a machine learning algorithm that trained on customer support conversations on Twitter. It also detects how tones progress throughout conversations, and offers suggestions on when agents should be more sympathetic, polite, or excited during an interaction.” Watson Tone analyzer assists the chatbot to take better informed decision.
Chatbot design approaches for fashion E-commerce: an interdisciplinary review
Published in International Journal of Fashion Design, Technology and Education, 2022
A. R. D. B. Landim, A. M. Pereira, T. Vieira, E. de B. Costa, J. A. B. Moura, V. Wanick, Eirini Bazaki
When user interactions are enacted through buttons or multiple-choice interfaces, the expressed intents and/or other relevant information may be straightforwardly comprehended by the chatbot. However, when human natural language contained in unstructured data is allowed in the form of text (possibly extracted from speech), specific algorithms for Natural Language Understanding (NLU) must be employed. Rule-based algorithms employ handcrafted hard rules by using, for instance, the Artificial Intelligence Mark-up Language (AIML). Classical ML algorithms in general combine the extraction of handcrafted features from unstructured textual data, such as n-gram counts or the Term Frequency–Inverse Document Frequency (TF-IDF) statistical measure. DL algorithms present state-of-the-art performance by employing deep neural network architectures for sequence processing, which mainly include variations of CNN and RNN to recognise complex patterns from data.
Recent advances in artificial intelligence for video production system
Published in Enterprise Information Systems, 2023
YuFeng Huang, ShiJuan Lv, Kuo-Kun Tseng, Pin-Jen Tseng, Xin Xie, Regina Fang-Ying Lin
In the field of text generation, the main general technologies include RNN, LSTM and Transformer. Generally speaking, NLP has two branches: natural language generation (NLG) and natural language understanding (NLU). Natural language generation is a more official academic term. In fact, the common ‘dialogue system generation’, ‘text automatic summarisation’, ‘intelligent writing’, etc., all contain NLG related technologies. In deep learning, recurrent neural networks (RNN) is a series of neural networks that are good at learning from sequence data, and long short-term memory (LSTM) is recurrent neural network that is applied in practice.