CitySpec: An Intelligent Assistant System for Requirement Specification in Smart Cities

Abstract

An increasing number of monitoring systems have been developed in smart cities to ensure that a city's realtime operations satisfy safety and performance requirements. However, many existing city requirements are written in English with missing, inaccurate, or ambiguous information. There is a high demand for assisting city policy makers in converting human-specified requirements to machine-understandable formal specifications for monitoring systems. To tackle this limitation, we build CitySpec, the first intelligent assistant system for requirement specification in smart cities. To create CitySpec, we first collect over 1,500 real-world city requirements across different domains from over 100 cities and extract city-specific knowledge to generate a dataset of city vocabulary with 3,061 words. We also build a translation model and enhance it through requirement synthesis and develop a novel online learning framework with validation under uncertainty. The evaluation results on real-world city requirements show that CitySpec increases the sentence-level accuracy of requirement specification from 59.02% to 86.64%, and has strong adaptability to a new city and a new domain (e.g., F1 score for requirements in Seattle increases from 77.6% to 93.75% with online learning).

CitySpec with Shield: A Secure Intelligent Assistant for Requirement Formalization

“…The evaluation results show that CitySpec is effective on supporting policy makers accurately writing and refining their requirements. This paper is an extension of [1]. We extend with the following new contributions.…”

“…Despite that extensive research efforts have been spent toward improving the expressiveness of specification languages and efficiency of the monitoring algorithms, the research challenge of how to convert human-specified requirements to machine-understandable formal specifications has received only scant attention. Moreover, our study (see Section 2) on over 1,500 real-world city requirements across different domains 1 shows that, first, existing city requirements are often defined with missing information or ambiguous description, e.g., no location information, using words like nearby, or close to. They are not precise enough to be converted to a formal specification or monitored in a city directly without clarifications by policy makers.…”

Auto311: A Confidence-Guided Automated System for Non-emergency Calls

“…Question Answering and Large Language Models. In recent years, advanced question-answering systems have evolved across various scenarios (Chen et al 2023(Chen et al , 2022bDiefenbach et al 2018). Black-box abstractive QA systems like mBART and T5 (Chipman et al 2022;Raffel et al 2019) lack output control.…”

An Intelligent Assistant for Converting City Requirements to Formal Specification

“…In this demonstration, we show CitySpec [1], an intelligent assistant system for smart city requirement specifications that bridges the gap between English requirements and formal specifications while also efficiently dealing with missing, inaccurate, or ambiguous information. CitySpec is designed to assist city policy makers in precisely filling out English city requirements using an intelligent interface, and then automatically converting them to formal specifications.…”