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 118 for “"human evaluation"”.
-
Recycling texts: human evaluation of example-based machine translation subtitles for DVD
… this prospective profiling phase, we also elicit human judgements (through a combined questionnaire and interview) on the quality of the corpus data and on the reusability in new contexts of the TL subtitles. The intelligibility and acceptability of EBMT-produced subtitles (dependent variables) …
-
The impact of paratext on readers of generative literature: Human evaluation of generated text
Generative literature has been emerging as a creative form since before the digital era, with one of its defining characteristics being the particular circumstances of its production. Since 2013, the National Novel Generation Month (NaNoGenMo) challenge has been one of the most prolific organized …
-
Evaluating the Evaluators: Subjective Bias and Consistency in Human Evaluation of Natural Language Generation
… Generation (NLG) community relies on shared evaluation techniques to understand progress in the field. Based on an analysis of papers published over 10 years (from 2008 to 2018) in NLG-specific conferences and on an observational study, this thesis identifies shortcomings with existing …
-
Human evaluation and statistical analyses on machine reading comprehension, question generation and open-domain dialogue
Evaluation is a critical element in the development process of many natural language based systems. In this thesis, we will present critical analyses of standard evaluation methodologies applied in the following Natural Language Processing (NLP) domains: machine reading comprehension (MRC), …
-
Enhancing neural machine translation of low-resource languages: corpus development, human evaluation and explainable AI architectures
… from the LoResMT2021 Shared Task. A subsequent human evaluation using the multidimensional quality metrics error taxonomy showcased the superior performance of the Transformer system in reducing both accuracy and fluency errors compared to an RNN-based counterpart. Furthermore, this thesis …
-
Response Generation in Longitudinal Dialogues
… the most challenging type of conversations for human-machine dialogue systems. LDs include the recollections of events, personal thoughts, and emotions specific to each individual in a sparse sequence of dialogue sessions. Dialogue systems designed for LDs should uniquely interact with the users …
-
SELF-SUPERVISED MODELING FOR OPEN-DOMAIN DIALOGUE EVALUATION
… complex task. The current gold standard is human evaluation, but its high cost, time-intensiveness, lack of scalability, and irreproducibility often necessitate the use of automatic evaluation as an alternative. In this thesis, we tackle three important challenges in the field of automatic …
-
Reducing Human Labor Cost in Deep Learning for Natural Language Processing
… the NLP models also requires expansive human evaluation. This dissertation focuses on reducing such human labor costs in deep learning for NLP. We develop novel frameworks for training deep learning models with limited/noisy annotation and a novel framework for estimating human …
-
Cyber Deception against Adversarial Reconnaissance in Enterprise Network using Semi-Indistinguishable Honeypot
… by adversaries, and (2) the absence of effective human evaluation in current literature to measure deception capability. While previous works explored honeypot-based deception strategies, very few have deployed their systems in real life, considered detection avoidance by attackers, and evaluated …
-
Component Study of Co -Word Analysis
The results of human evaluation and statistical analyses show that both the strength index component and input term component affect the output from co-word analysis in significant ways. The resulting maps from the inclusion index test are much more general than those from the equivalence index. …
-
Personalo tarnybos automatinio asistento, panaudojančio RAG ir padedančio darbuotojams susipažinti su įmonės procedūromis, projektavimas ir vertinimas /
… an average reply time of 15 seconds, and top human evaluation scores for clarity and relevance.
-
Generation of Semantically Consistent Text and Its Evaluation
… formalisation of semantic errors and robust evaluation methods. First, it synthesises prior work on error annotation to develop a task-agnostic consensus taxonomy for semantic errors, distinguishing omission, addition, and substitution errors. Second, it applies this taxonomy in human …
-
Multidimensional opinion mining from social data
… approaches in the context of an extrinsic human evaluation conducted in a real-world context: the Malta Government Budget, where five social opinion dimensions are taken into consideration, namely subjectivity, sentiment polarity, emotion, irony, and sarcasm. This human evaluation …
-
Neural generation of textual summaries from knowledge base triples
… Using methods for both automatic and human evaluation, we demonstrated that our technique is capable of scaling to domains with challenging vocabulary sizes of over 400k words.<br/><br/>Given the promising results of our approach in biographies, we explore its applicability in the …
-
Methods to Evaluate and Improve the Quality of Text Summarization
… summarization is highly essential. Summarization evaluation is still challenging as there are no ideal evaluation methods. The automatic method is fast, inexpensive, and can handle large volumes of data without human intervention. However, some studies demonstrate that existing automatic …
-
The integration of machine translation and translation memory
… effort. We perform both automatic and human evaluation on these models. When measured against the consensus of human judgement, the recommendation model obtains 0.91 precision at 0.93 recall, and the reranking model obtains 0.86 precision at 0.59 recall. The high precision of these …
-
Lactose Hydrolysis by Fungal and Yeast Lactase: Influence on Freezing Point and Dipping Characteristics of Ice Cream
… methods could be used as an alternative to human testing of dippability. In the first experiment, ice cream mixes were treated with lactase (EC 3.2.1.23) to cause 0 to 83% lactose hydrolysis. Lactose hydrolysis decreased the freezing point from -1.63oC in the control (0% hydrolysis) to …
-
Multimodal Film Music Dataset
… our extraction and annotation procedures through human evaluation of cue detection and music-feature recognition. Exploratory analyses illustrate how orchestration varies by genre, how soundtrack emotion tracks narrate mood, and how the presence of dialogue shapes musical impact. By sharing both …
-
Text summarization using transfer learnin: Extractive and abstractive summarization using BERT and GPT-2 on news and podcast data
… to state-of-the-art. Besides the quantitative evaluation, we also perform a qualitative investigation in the form of a human evaluation, along with inspection of the trained model that demonstrates that it learns reasonable abstractions.
-
Contextual Scripture Recommendation for Writers
… and compare automatic metrics with a small human evaluation.
Page 1 of 6