The widespread usage of rdbms in the commercial world has played a pivotal role in the offering of database systems course in major universities. A key challenge encountered by learners taking such a course is the topic of query optimization. The query optimization process produces a query execution plan (qep), which represents an execution strategy for an sql query. Unfortunately, in practice, it is often difficult for a learner to comprehend query execution strategies by perusing vendor-specific qeps, hindering her learning process. In this paper, we present a novel, end-to-end, generic sys- tem called lantern that generates a natural language description of a qep to enhance its understanding. It takes as input an sql query and its qep, and generates a natural language description of the execution strategy deployed by the underlying rdbms. Specifi- cally, it deploys a declarative framework called pool that enables subject matter experts to efficiently create and maintain natural language descriptions of physical operators used in qeps. A rule- based framework called rule-lantern is proposed that exploits pool to generate natural language descriptions of qeps. Despite the high accuracy of rule-lantern, our engagement with learners reveal that consistent with existing psychology theories perusing such rule-based descriptions lead to boredom due to repetitive state- ments across different qeps. To address this issue, we present a novel deep learning-based language generation framework called neural-lantern that infuses language variability in the gener- ated description by exploiting a set of paraphrasing tools and word embedding. Our experimental study with real learners shows the effectiveness of lantern in facilitating comprehension of qeps.

@inproceedings{wang-et-al-sigmod-21, abstract = {The widespread usage of rdbms in the commercial world has played a pivotal role in the offering of database systems course in major universities. A key challenge encountered by learners taking such a course is the topic of query optimization. The query optimization process produces a query execution plan (qep), which represents an execution strategy for an sql query. Unfortunately, in practice, it is often difficult for a learner to comprehend query execution strategies by perusing vendor-specific qeps, hindering her learning process. In this paper, we present a novel, end-to-end, generic sys- tem called lantern that generates a natural language description of a qep to enhance its understanding. It takes as input an sql query and its qep, and generates a natural language description of the execution strategy deployed by the underlying rdbms. Specifi- cally, it deploys a declarative framework called pool that enables subject matter experts to efficiently create and maintain natural language descriptions of physical operators used in qeps. A rule- based framework called rule-lantern is proposed that exploits pool to generate natural language descriptions of qeps. Despite the high accuracy of rule-lantern, our engagement with learners reveal that consistent with existing psychology theories perusing such rule-based descriptions lead to boredom due to repetitive state- ments across different qeps. To address this issue, we present a novel deep learning-based language generation framework called neural-lantern that infuses language variability in the gener- ated description by exploiting a set of paraphrasing tools and word embedding. Our experimental study with real learners shows the effectiveness of lantern in facilitating comprehension of qeps.}, address = {Xi'an, Shaanxi, China}, author = {Weiguo Wang and Sourav S Bhowmick and Hui Li and Shafiq Joty and Siyuan Liu}, booktitle = {Proceedings of 2021 ACM SIGMOD International Conference on Management of Data}, month = {June}, pages = {x -- x}, publisher = {ACM}, series = {SIGMOD'21}, title = {Towards Enhancing Database Education: Natural Language Generation Meets Query Execution Plans}, url = {}, year = {2021} }