Natural Language Generation Algorithms Research Articles

A Natural Language Generation Algorithm for Greek by Using Hole Semantics and a Systemic Grammatical Formalism

This work is about the progress of previous related work based on an experiment to improve the intelligence of robotic systems, with the aim of achieving more linguistic communication capabilities between humans and robots. In this paper, the authors attempt an algorithmic approach to natural language generation through hole semantics and by applying the OMAS-III computational model as a grammatical formalism. In the original work, a technical language is used, while in the later works, this has been replaced by a limited Greek natural language dictionary. This particular effort was made to give the evolving system the ability to ask questions as well as the authors developed an initial dialogue system using these techniques. The results show that the use of these techniques the authors apply can give us a more sophisticated dialogue system in the future.

Intelligent visualization and visual analytics

Intelligent visualization and visual analytics

Artificial intelligence versus Maya Angelou: Experimental evidence that people cannot differentiate AI-generated from human-written poetry

The release of openly available, robust natural language generation algorithms (NLG) has spurred much public attention and debate. One reason lies in the algorithms' purported ability to generate humanlike text across various domains. Empirical evidence using incentivized tasks to assess whether people (a) can distinguish and (b) prefer algorithm-generated versus human-written text is lacking. We conducted two experiments assessing behavioral reactions to the state-of-the-art Natural Language Generation algorithm GPT-2 (Ntotal = 830). Using the identical starting lines of human poems, GPT-2 produced samples of poems. From these samples, either a random poem was chosen (Human-out-of-theloop) or the best one was selected (Human-in-the-loop) and in turn matched with a human-written poem. In a new incentivized version of the Turing Test, participants failed to reliably detect the algorithmicallygenerated poems in the Human-in-the-loop treatment, yet succeeded in the Human-out-of-the-loop treatment. Further, people reveal a slight aversion to algorithm-generated poetry, independent on whether participants were informed about the algorithmic origin of the poem (Transparency) or not (Opacity). We discuss what these results convey about the performance of NLG algorithms to produce human-like text and propose methodologies to study such learning algorithms in human-agent experimental settings.

Personal research idea recommendation using research trends and a hierarchical topic model

In the era of rapid technological advance, it is an important task for all researchers to keep up with trends when performing research. How to efficiently find suitable research topics while the number of papers is increasing rapidly is worthwhile to explore. To solve such problems, some researchers attempted to find research ideas by topic detection and tracking methods. However, these methods do not consider the users’ background knowledge and preferences, and they express a topic with general keywords, which does not effectively help researchers to develop new research ideas. Existing studies support that the title expresses the research idea the best. This study adapts this concept to propose an automatic title generation method that combines personalized recommendation methods and topic trend analysis methods to achieve this task. First, it uses hierarchical latent tree analysis to find the users’ interests for a topic structure and its representative keywords hidden in the existing research. Second, the interesting topic trends, popularity and user preferences in a hybrid recommendation method are considered. Finally, a natural language generation algorithm that is suitable for the titles of academic papers converts the original recommended-keywords into fluent title sentences that are designed for the users. Experiments have found that adding Google Trend indicators and personal factors can improve the performance of topic recommendations. The automatic title generation method using template-based and statistical information methods leads to excellent performances in both grammatical correctness and semantic expression. Moreover, for the users, the title is indeed more inspirational than the simple keywords for users to develop new research ideas.

Do speakers articulate over-described modifiers differently from modifiers that are required by context? Implications for models of reference production

Studies have shown that speakers often include unnecessary modifiers when producing referential expressions, which is contrary to the Maxim of Quantity. In this study, we examined the production of referring expressions (e.g. the red triangle) that contained an over-described (or redundant) pre-nominal adjective modifier. These expressions were compared to similar expressions that were uttered in a context that made the modifier necessary for unique referent identification. Our hypothesis was that speakers articulate over-described modifiers differently from those used to distinguish contrasting objects. Results showed that over-described modifiers were significantly shorter in duration than modifiers used to distinguish two objects. Conclusions focus on how these acoustic differences can be modelled by Natural Language Generation algorithms, such as the Incremental Algorithm, in combination with probabilistic prosodic reduction.