Elasticsearch (ES) is a distributed RESTful search engine optimized for large-scale and long-text search scenarios. Recent research on text-to-Query has explored using large language models (LLMs) to convert user query intent to executable code, making it an increasingly popular research topic. To our knowledge, we are the first to introduce the novel semantic parsing task text-to-ES. To bridge the gap between LLM and ES, in detail, we leverage LLMs and employ domain experts to generate ES query bodies, which are Domain-Specific Language (DSL), along with the corresponding post-processing code to support multi-index ES queries. Consequently, we propose the text-to-ES benchmark that consists of two datasets: Large Elasticsearch Dataset (LED), containing 26,207 text-ES pairs derived from a 224.9GB schema-free database, and ElasticSearch (BirdES)with 10,926 pairs sourced from the Bird dataset on a 33.4GB schema-fixed database. Compared with fourteen advanced LLMs and six code-based LLMs, the model we trained outperformed DeepSeek-R1 by 15.64% on the LED dataset, setting a new state-of-the-art, and achieved 78% of DeepSeek-R1’s performance on the BirdES dataset. Additionally, we provide in-depth experimental analyses and suggest future research directions for this task. Our datasets are available at https://huggingface.co/datasets/Barry1915/Text-to-ES.

Distracted driving increases the risk of driving errors and accidents, and it has become the main factor leading to accidents. Currently, a large amount of information is displayed on in-vehicle panels, which possibly distract drivers’ attention, diverting their gaze from the road. To minimize the negative impact of distracted driving, this study sought to optimize the design of icons presented on vehicle screens. We investigated the effect of icon size, icon position, driving state, and age on drivers’ perception time, driving performance, and subjective evaluations. A total of 42 drivers (26 men and 20 women) between the ages of 22 and 70 years participated in the study, completing a simulated driving task and a secondary visual task. The results demonstrated significant main effects of icon size, driving state, icon position, and interaction effects. Participants spent a longer time perceiving smaller icons on the center control panel in the driving state. As the icon size increases, perception time decreases at a reduced speed. During the static state, perception time leveled out from 44′ while during the driving state, perception time leveled out from 80′. On the instrument panel, perception time leveled out from 60′ while on the center control panel, perception time leveled out from 80′. Subjective evaluations supported behavioral data on the effect of icon size. Our results indicated the importance of strategic design for icons on vehicle screens, which could reduce the visual workload, then reduce the risk of traffic accidents. According to the results, enterprises or designers could target the design of icon size according to different types of in-vehicle panels or screens.