System and method for facilitating efficient indexing in a database system
AssigneeBeijing OceanBase Technology Co., Ltd.
InventorZhiyong Jiang
Life of this patent
Estimating the term…
Abstract
Embodiments described herein provide a system for facilitating efficient indexing in a database. The system can maintain a data table in the database. The data table includes a first column for structured data and a second column for unstructured text. A row of the data table can include a first piece of data in the first column and a second piece of data in the second column. During operation, the system performs word segmentation on the second piece of data to generate a first set of semantically meaningful word segments. The system then generates a set of index entries that corresponds to the row. A respective index entry includes an index generated from the first piece of data and a corresponding word segment of the first set of semantically meaningful word segments.
Claims20
A method for facilitating efficient indexing in a database, comprising:
- maintaining, by a database system, a data table in the database, wherein the data table includes a first column for structured data supported by the database system and a second column for unstructured text, and wherein a first row of the data table includes a first piece of data in the first column and a second piece of data in the second column;
- in response to the first row being generated, performing word segmentation on the second piece of data to generate a first set of semantically meaningful word segments, wherein a respective semantically meaningful word segment comprises a combination of one or more phrases from unstructured text in the second piece of data;
- generating, in an index table of the database, a set of index entries that points to the first row, wherein each index entry of the set of index entries comprises an index generated using the first piece of data stored in the first column of the first row, and wherein a respective index entry of the set of index entries comprises a corresponding word segment of the first set of semantically meaningful word segments;
- and
- in response to receiving a query to the database for the unstructured text, looking up, by the database system, a response to the query based on the index table.
A non-transitory computer-readable storage medium storing instructions that when executed by a computer, cause the computer to perform a method for facilitating efficient indexing in a database, the method comprising:
- maintaining a data table in the database, wherein the data table includes a first column for structured data and a second column for unstructured text, and wherein a first row of the data table includes a first piece of data in the first column and a second piece of data in the second column;
- in response to the first row being generated, performing word segmentation on the second piece of data to generate a first set of semantically meaningful word segments, wherein a respective semantically meaningful word segment comprises a combination of one or more phrases from unstructured text in the second piece of data;
- generating, in an index table of the database, a set of index entries that points to the first row, wherein each index entry of the set of index entries comprises an index generated using the first piece of data stored in the first column of the first row, and wherein a respective index entry of the set of index entries comprises a corresponding word segment of the first set of semantically meaningful word segments;
- and
- in response to receiving a query to the database for the unstructured text, looking up, by the database system, a response to the query based on the index table.
Dependencies are inferred from the claim text.