Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 24 for “"intent classification"”.
-
Intent Classification during Human-Robot Contact
… is a need for robots to respond appropriately to intentional contact during interactive and collaborative tasks. This thesis proposes a classification-based approach for differentiating between several intentional contact types, accidental contact, and no-contact situations. A dataset is de- …
-
Enhancing Intent Classification via Zero-shot and Few-shot ChatGPT Prompting Engineering: Generating training data or directly detecting intents?
… (user utterances) για συγκεκριμένες προθέσεις (intents), οι οποίες φράσεις μπορούν να αξιοποιηθούν για την εκπαίδευση NLU μοντέλων. Στην πράξη, ωστόσο, έχει παρατηρηθεί ότι, όταν αυτές οι τεχνητά γεννημένες φράσεις χρησιμοποιούνται ως σύνολο εκπαίδευσης, οδηγούν σε μοντέλα μη ανταγωνιστικής …
-
NLP sistemų analizės programinės įrangos sudarymas ir testavimas /
… system, supporting REST API, for evaluating intent classification and sentiment analysis characteristics using cluster analysis.
-
Deep Generative and Explainable Learning Frameworks for Intent Recognition in Naval Domain
Accurate recognition of vessel intent supports maritime situational awareness, autonomous navigation, and tactical decision support. Three obstacles limit progress: the scarcity of labeled hostile-encounter data, noisy and intermittent sensor streams, and the inability of data-driven models to …
-
Query Expansion Techniques for Enterprise Search
… recognition, alphanumeric term identification, intent classification, collection enrichment, and word vectors were implemented and studied using real enterprise data. They were evaluated against a test set of queries developed using relevance survey results from multiple users, using standard …
-
Enriching Scientific Paper Embeddings with Citation Context
… I present a novel method for incorporating intent information into scientific paper embeddings through edge-weighting and concatenation of per-intent node2vec embeddings. Furthermore, I suggest that a hybrid approach, including both text and network data to generate embed- dings can take …
-
Neural Enhancement Strategies for Robust Speech Processing
… improve the performance of back-end speech-based classification tasks in noisy conditions. In detail, we propose a pipeline that integrates the Wave-U-Net (later this model is modified to the Dilated Encoder Wave-U-Net) as a pre-processing stage for noise elimination with a temporal convolution …
-
Adaptive open-set active learning with distance-based out-of-distribution detection for robust task-oriented dialog systems
… learning (AL) have become salient for user intent classification performance in task-oriented dialog systems (TODS). In realistic settings, however, traditional AL techniques often fail to efficiently select targeted in-distribution (IND) data when encountering newly acquired outof- …
-
Virtual Assistant Design for Water Systems Operation
… model. Additionally, we explore the design of intent classifiers, which involves more challenges than a traditional classifier due to the small ratio of text length compared to the number of classes. In our design, we incorporate the results of entity recognition, produced from previous layers …
-
AI-Powered chatbot for hotel reservation systems /
… (LLMs). First, user messages are processed by an intent-classification step which applies both TF-IDF model and linear model. Depending on what the user wants to do (usually, booking, canceling, or information), the classifier sends the chat to the appropriate section of the system. Booking, when …
-
Σχεδίαση και Υλοποίηση Διαλογικού Συστήματος Υποστήριξης για Άγχος και Συναφείς Δυσκολίες με Τεχνικές Μηχανικής Μάθησης
… pipeline δύο σταδίων: (α) αναγνώριση πρόθεσης (intent classification) με επιβλεπόμενη μηχανική μάθηση και (β) ανάκτηση απάντησης (retrieval) από επιμελημένη βάση γνώσης ερωτήσεων-απαντήσεων (Q/A), όπου κάθε απάντηση συνοδεύεται από πηγή. Για την αναγνώριση προθέσεων υλοποιήθηκε baseline μοντέλο …
-
Grėsmių socialinėse medijose stebėjimo įrankis /
… analysis of communication content related to intent interpretation is typically offered only as an auxiliary feature. The implemented system integrates data collection, natural language processing, threat profile application, risk assessment, and incident management. It collects messages from …
-
Multimodal intent recognition for natural human-robotic interaction
… for topic spotting be used to identify spoken intent in unconstrained natural speech? Can gesture recognition systems based on statistical speech recognition techniques be used to bridge the gap between physical movements and recognition of gestural intent? How can speech and gesture be …
-
Low-Resource Natural Language Understanding in Task-Oriented Dialogue
… the user's input to understand the user's needs (intent) and corresponding relevant information (slots). This process is performed by a Natural Language Understanding (NLU) component, which maps the text utterance into a semantic frame representation, involving two subtasks: intent classification …
-
Understanding user intents in online health forums
… extract useful information. Understanding the intents of people who post on these forums would enable the retrieval of relevant information from existing threads which would in turn allow users to more effectively find answers to their medical needs. In this paper, we derive a taxonomy of …
-
Design and Development of a Chatbot as an Alternative Web-browser for those with Severe Motor Impairments
… technique was developed. This method enhanced intent classification accuracy by 4.39% (from 75.86% to 80.25%) without compromising user privacy or requiring large-scale data collection from vulnerable populations. Using an optimized AR-based eye-gaze chatbot, this dissertation lays the …
-
Learning speech embeddings for speaker adaptation and speech understanding
… We show that this technique increases the intent classification accuracy as compared to a speech-only system. The second SLU problem is the label imbalance problem in the datasets, which is also related to fairness since a model trained on skewed data usually leads to biased results. To …
-
Multilingual Spoken Language Understanding: Efficient Speech Dataset Collection, Architectural Exploration, and Zero-Shot SLU
… is a key technology for interpreting a speaker’s intention through intent classification (IC) and extracting relevant information via slot filling (SF). For example, when a user asks, “What is the weather like in London tomorrow?”, the intent is ‘Get Weather’ and the slots are ‘location: London’ …
-
Automatic Activity Identification in Unstructured Communication Artifacts
… of time finding subtasks such as actions or intents in order to track activities. The automatic identification of these actions or intents from email or conversation can increase workplace productivity. Emails are widely used in this research topic, while the use of conversations from …
-
A retrospective analysis of pharmaceutical and non-pharmaceutical substance-related coronial deaths in South Australia over 20 years
… pharmaceutical substance-related deaths were unintentional, in stark contrast to non-pharmaceutical substance-related deaths where more than 90% of cases were suicides. This finding highlights the importance of providing consistent and comprehensive analysis of not only pharmaceutical …
Page 1 of 2