All Products
Search
Document Center

Comparison of General-purpose Edition and Industry-specific Enhanced Edition

Last Updated: Sep 09, 2021

Industry-specific Enhanced Edition

The search performance determines the gross merchandise volume (GMV) in E-commerce. To achieve better development, E-commerce enterprises must endeavor to achieve good search performance. OpenSearch provides exclusive solutions for E-commerce enterprises in vertical industries based on the industrial experience accumulated over the years by Alibaba. OpenSearch adopts the Natural Language Processing (NLP) technology developed by Alibaba DAMO Academy to resolve the tough issues and satisfy the requirements in the industries. OpenSearch allows E-commerce enterprises to perform query analysis based on their dedicated requirements and provides built-in sort expressions and algorithms to satisfy the industrial needs. OpenSearch is easy to use and can be configured with a few clicks. This improves access efficiency and enables better search performance for enterprises.

Feature comparison

Feature

General-purpose Edition

Industry-specific Enhanced Edition

One-stop configuration

After an application is created, you need to create and configure query analysis policies, sort policies, and drop-down suggestion models manually.

This edition provides the capabilities and features that are required for common search scenarios in E-commerce. It also provides templates for you to configure application schemas and index schemas with a few clicks. This way, new users can learn how to use this service with ease.

Query analysis

This edition provides common query rewriting capabilities for general industries. The capabilities include synonym matching, stop word filtering, spelling correction, term weight analysis, and category prediction.

This edition provides enhanced analyzers and query analysis features for the general industries and common verticals of E-commerce, such as fashion, automobile, and food & drink industries. This edition is developed based on the characteristics of goods and users in E-commerce. It can create more precise indexes and better identify user needs. Therefore, this edition has better performance than General-purpose Edition.

Sort policy

After an application is created, you need to configure the sort policy and perform debugging based on the business scenarios manually.

In addition to templates for application schemas and index schemas, this edition also provides common rough sort expressions and fine sort expressions. The templates and expressions can satisfy the sort requirements in most E-commerce scenarios.

Feature iteration

This edition regularly updates the default system dictionaries, such as the dictionaries for the analyzer and query analysis features.

This edition is constantly updated in line with the changes of words and services in E-commerce. It keeps optimizing the existing features of analysis and query analysis to adapt to industrial changes.

Comparison of query analysis performance

Compared with General-purpose Edition, Industry-specific Enhanced Edition provides better query analysis performance. For example, Industry-specific Enhanced Edition corrects the invalid entries in the existing dictionary of General-purpose Edition. Industry-specific Enhanced Edition enriches the existing dictionary based on the characteristics of words used in various E-commerce industries. The following tables compare the performance of the two editions in different query analysis capabilities.

Space-based analysis

Query

General-purpose Edition

Industry-specific Enhanced Edition

汉本萃 纽崔莱

汉 本 萃 纽崔莱

汉本萃 纽崔莱

小金管遮瑕膏

小金 管 遮瑕 膏

小 金管 遮瑕 膏

苏泊尔空气炸锅

苏泊尔 空气 炸 锅

苏泊尔 空气 炸锅

寻梦环游记吉他

寻梦 环 游记 吉他

寻梦 环游记 吉他

炸街西装套装

炸 街 西装 套装

炸街 西装 套装

喜戈洗发水

喜 戈 洗发 水

喜戈 洗发 水

橘朵单色眼影

橘 朵 单色 眼影

橘朵 单色 眼影

Entity identification

Query

General-purpose Edition

Industry-specific Enhanced Edition

苹果 11 壳 女款

苹果: the category of the commodity. 11: a number. 壳 a common term. 女款: the intended customers of the commodity.

苹果: the brand name of the commodity. 11: a number. 壳: the category of the commodity. 女款: the intended customers of the commodity.

汉本萃 纽崔莱

汉本萃: a common term. 纽崔莱: the brand name of the commodity.

汉本萃: the brand name of the commodity. 纽崔莱: the brand name of the commodity.

蔬菜苗

蔬菜苗: a new term

蔬菜苗: the category of the commodity

美瞳

美瞳: a common term

美瞳: the category of the commodity

Synonym

Query

General-purpose Edition

Industry-specific Enhanced Edition

阔腿裤女2021年新款

阔腿裤 女 2021年 新款

阔腿裤(腿裤 阔脚裤 甩裤) 女 2021年 新款

汉本萃 纽崔莱

汉本萃 纽崔莱

汉本萃 纽崔莱(Nutrilite)

运动鞋女跑步鞋

运动鞋 女 跑步鞋(跑鞋)

运动鞋 女 跑步鞋(跑鞋 慢跑鞋 健步鞋)

苏泊尔空气炸锅

苏泊尔 空气 炸锅

苏泊尔(supor 苏泊儿 苏伯尔) 空气 炸锅

洗碗机西门子

洗碗机 西门子

洗碗机 西门子(siemens)

Spelling correction

Query

General-purpose Edition

Industry-specific Enhanced Edition

无刚圈文胸

无刚圈文胸

无钢圈文胸

putizhu

putizhu

菩提珠

baisiwa

baisiwa

白丝袜

红色baobao 女包

红色baobao 女包

红色包包 女包

sushe 冰箱

sushe 冰箱

宿舍 冰箱

chuanglian欧美

chuanglian欧美

窗帘欧美

孕妇装dama

孕妇装dama

孕妇装大码

u型枕 nami

u型枕 nami

u型枕 纳米

按磨器 腰部

按磨器 腰部

Massor Waist

yami

yami

雅蜜

yingyang

yingyang

营养

jiankang

jiankang

健康

呀高

呀高

牙膏

Upgrade from General-purpose Edition to Industry-specific Enhanced Edition

If you are an old user of OpenSearch, OpenSearch allows you to upgrade an application from General-purpose Edition to Industry-specific Enhanced Edition. Perform the following steps:

1. On the Applications page, select an application of the General-purpose Edition, click More in the Actions column, and then select Upgrade to Industry-specific Enhanced Edition to go to the buy page.

1

23

2. After the application is upgraded to Industry-specific Enhanced Edition, click Adapt to Industry-specific Enhanced Edition in the Actions column to modify the application.

4

3. Modify the application by performing the following steps: select template features, define the application schema, define the index schema, and configure data sources.

Select template features:

The template features are classified into the following categories: Query Analysis, Sort Policy, and Other Features.

5

Define the application schema:

The application schema can be automatically created.

6

Note: Field tags are used to generate the index schema, query analysis rules, and sort policies for the application in Industry-specific Enhanced Edition. Make sure that the field names associated with the field tags are specified in the application schema. For more information, see the preceding figure.

Define the index schema:

7

Note: Index tags are used to generate query analysis rules and sort policies. Make sure that the index names associated with the index tags are specified in the index schema. For more information, see the preceding figure.

4. Configure data sources. The following figure shows how to configure an ApsaraDB RDS for MySQL data source.

8

5. Sort the search results.

99

Note: If the fields referenced in a sort expression are changed, edit or delete the invalid sort expression.

6. After the configurations are modified, determine whether to synchronize incremental data to the offline application and whether to automatically publish the offline application after the configurations of the offline application are modified.

10
  • You may determine not to synchronize incremental data but to push full data to the offline application. When the offline application is being created, the incremental data that is pushed to the online application cannot be synchronized to the offline application.

  • You may determine not to automatically publish the modified offline application to online environments. In this case, click Publish on the Offline Application tab of the Application Details page to manually publish the offline application.

11

7. Test the application after the configurations are complete.

12

Usage notes

  • Applications can be upgraded from General-purpose Edition to Industry-specific Enhanced Edition but cannot be downgraded from Industry-specific Enhanced Edition to General-purpose Edition.

  • Only exclusive applications can be upgraded to Industry-specific Enhanced Edition.

  • Make sure that the field names associated with field tags are specified in the application schema. Otherwise, an error message is prompted.

001