1.03 Natural language processing (NLP) and understanding

Natural Language Processing, commonly abbreviated as NLP, is a truly fascinating and rapidly advancing subfield of Artificial Intelligence. At its core, NLP is dedicated to enabling computers and software systems to effectively understand, accurately interpret, appropriately respond to, and even convincingly generate human language – encompassing both spoken words and written text. NLP is the sophisticated technology that underpins many applications we use daily: it allows your smartphone’s virtual assistant to comprehend your voice commands when you ask for directions or to set a reminder; it powers the intelligent chatbots on websites that attempt to answer your customer service questions in real-time; and it enables powerful translation tools to convert entire documents or websites from one human language to another with increasing accuracy.

But how does NLP actually work, given the inherent complexity of human language? Human languages are rich with nuance, ambiguity, slang, metaphors, and context-dependent meanings, all of which make them challenging for computers, which prefer precision and unambiguous instructions. Computers don’t “understand” language in the deeply intuitive and contextual way humans do. NLP addresses this challenge by employing a diverse array of computational techniques and algorithms to break down human language into more manageable components that computer systems can process and analyse. This intricate process might involve:

• Tokenisation: Dividing text into smaller units like words or sentences.
• Part-of-speech tagging: Identifying the grammatical role of each word (e.g., noun, verb, adjective, adverb).
• Parsing: Analysing the grammatical structure of sentences to understand the relationships between words.
• Named entity recognition (NER): Identifying and categorising key entities in text, such as names of people, places, organisations, dates, and monetary values.
• Sentiment analysis: Attempting to determine the emotional tone or subjective opinion expressed in a piece of text (e.g., positive, negative, neutral).
• Topic modelling: Discovering abstract topics that occur in a collection of documents.

Imagine a computer system tasked with processing an email application for a job. NLP techniques would help it recognise distinct sections: the greeting (“Dear Hiring Manager,”), the main body of the message detailing the applicant’s relevant skills and professional experience, and the closing statement (“Sincerely, [Applicant Name]”). This sophisticated ability to dissect and “make sense” of human language structure and content makes communication between people and computers significantly smoother, more intuitive, and more effective.

In the realm of adult education, NLP offers a multitude of valuable ways to support both learners in their educational journeys and educators in their teaching practices. For instance:

• Making information more accessible: Many adults returning to education, or those who find reading challenging, may struggle with lengthy, dense, or technically complex texts. NLP-powered tools can provide invaluable assistance by automatically summarising long articles into concise key bullet points, highlighting the most important ideas and takeaways, or even rewriting the original content using simpler vocabulary, shorter sentences, and clearer structures, thereby enhancing comprehension.
• Providing on-demand learner support: In online courses or blended learning environments, chatbots equipped with sophisticated NLP capabilities can answer frequently asked questions from learners, 24/7. This provides immediate support, helps resolve common doubts or misunderstandings quickly, and is particularly beneficial for adult learners who often study at varied times due to work, family, or other commitments, enabling them to get help even outside of standard instructor hours.
• Assisting with various forms of communication: Speech recognition systems, which are a key application of NLP, can accurately convert spoken words – such as answers to questions, oral presentations, or dictated notes – into editable written text. This is a significant boon for learners who find typing difficult, slow, or physically challenging. Conversely, text-to-speech (TTS) systems, also driven by NLP, can read digital text aloud in natural-sounding voices, greatly benefiting learners with visual impairments, learning differences like dyslexia, or those who simply prefer auditory learning.
• Facilitating language learning and translation: NLP is the fundamental technology behind modern language translation tools. Adult learners can use these tools to understand educational materials written in a foreign language, or to get assistance in expressing themselves accurately and fluently when writing or speaking in a new language they are learning. This is exceptionally useful in diverse, multicultural learning groups or for international collaborative projects, such as the HER[AI]TAGE project which involves partners from different European countries.
• Improving writing quality: Even the everyday tools we take for granted, like spell checkers and grammar checkers embedded in word processors, email clients, and online platforms, rely heavily on NLP. These tools analyse the text as you type, identify potential spelling errors, grammatical mistakes, awkward phrasing, or stylistic inconsistencies, and suggest corrections or improvements, helping everyone to communicate more clearly and professionally.

NLP is also a critical enabling technology for many accessibility features that make learning environments more inclusive and equitable. By empowering computers to process, understand, and generate human language more effectively, NLP helps to break down communication barriers, cater to diverse learning preferences, and create more supportive, adaptable, and ultimately more successful learning experiences for adults from all backgrounds and with a wide range of individual needs.

PRACTICAL EXAMPLES

• A research group working on the HER[AI]TAGE project uses NLP tools to perform sentiment analysis on a large corpus of community feedback forms regarding a proposed cultural exhibition. This helps them quickly gauge public opinion, identify common concerns, and understand the overall emotional response to their plans.
• An adult learner is tasked with preparing a detailed report for their job but is not very confident or efficient with typing. They utilise their smartphone’s advanced voice input feature (a speech-to-text application powered by NLP) to dictate the entire report, including complex sentences and specialised terminology. The system then accurately converts their spoken words into a well-formatted written document, saving them considerable time and effort.
• An educator wishes to share a comprehensive and lengthy research article about local biodiversity and its cultural significance with a community group actively involved in the HER[AI]TAGE project. To make the information more digestible and save the community members valuable time, the educator uses a sophisticated AI summarisation tool (which employs NLP techniques like identifying key sentences and concepts) to create a short, clear, and easy-to-understand overview of the article’s main findings, conclusions, and recommendations.
• A learner whose first language is not English is enrolled in an online vocational training course. They frequently use an NLP-powered translation app integrated into their web browser to instantly translate difficult technical phrases or complex instructional sentences within the course materials. This not only aids their comprehension but also helps them build confidence in reading and actively participating in forum discussions.
• During an interactive online workshop for adult educators focusing on new digital pedagogy techniques, an NLP-driven chatbot is available in the virtual meeting room. It is programmed to answer common administrative and logistical questions (e.g., “Where can I download the presentation slides?” or “What is the schedule for tomorrow’s sessions?”). This efficient handling of routine queries frees up the human facilitator to concentrate on leading engaging discussions, responding to more complex pedagogical questions, and providing individualised support to participants.
• An adult learning centre is committed to inclusivity and makes all its course handbooks, policy documents, and online resources available in a digital format that is fully compatible with text-to-speech software. This NLP-driven feature allows learners with dyslexia, visual impairments, or other reading challenges to have the materials read aloud to them, ensuring they have equal access to information and making the overall learning environment more welcoming and supportive.
• A generative AI tool, heavily relying on its NLP capabilities to understand intent and generate appropriate language, assists an adult learner in drafting a polite, professional, and persuasive email to a potential mentor, inquiring about the possibility of an informational interview to learn more about their career field.