Private domain retail search hot word prediction and verification method based on social information insight

Abstract

The invention discloses a private domain retail search hot word prediction and verification method based on social information insight, and relates to the field of retail industry. The method comprises the following steps: newly creating a project: an administrator newly creating an insight project in a management background, inputting content information expected to be insight in a content definition module, and setting specific requirements of insight; insight project monitoring: executing the obtained content information which is expected to be insight through the monitoring module; and information aggregation: performing information aggregation on the information data acquired by the monitoring module and executing total data statistics after the drafted period is finished, and generating a data report for summarization. According to the method, the cost of acquiring the industry data and the brand data is reduced, the corresponding speed of the hotspot information is improved, and the timeliness and the popularity of the information are ensured by performing data monitoring by using the full-amount platform data; the monitoring content is optimized, and multi-channel and multi-angle hotspot information is obtained.

Description

technical field [0001] The invention belongs to the field of the retail industry, and in particular relates to a private domain retail search hot word prediction and verification method based on social information insights. Background technique [0002] In the existing retail industry, public opinion monitoring in the retail industry requires high cost of obtaining public opinion content, a large amount of data monitoring, and manual review of the content in the platform. Although most platforms have certain filtering mechanisms, However, the limited regulation mechanism cannot deal with all kinds of new words well, and often requires manual in-depth review of data sources to analyze some information, and the cost of obtaining information is very high; and the response speed to hot content is slow, existing In the current public opinion monitoring system, due to the defects of the above-mentioned filtering mechanism, new hotspots need to be settled for a period of time befor...

AI Technical Summary

Problems solved by technology

[0002] In the existing retail industry, public opinion monitoring in the retail industry requires high cost of obtaining public opinion content, a large amount of data monitoring, and manual review of the content in the platform. Although most platforms have certain filtering mechanisms, However, the limited regulation mechanism cannot deal with all kinds of new words well, and often requires manual in-depth review of data sources to analyze some information, and the cost of obtaining information is very high; and the response speed to hot content is slow, existing In the current public opinion monitoring system, due to the defects of the above-mentioned filtering mechanism, new hotspots need to be settled for a period of time before they have a chance to be captured by the rules, and since the information sources are all from the platform, the collected information is affected by the "hot spots" of the third-party platform. "search" around, unable to quickly respond to brands in this industry, which greatly weakens the brand's response speed to its own content; in the existing public opinion monitoring system, the perspective of public opinion information monitoring is single, and mainly negative information, such as Weibo, Less than a quarter of daily hot search content is positive information, and the proportion of positive information is too low. It is difficult for brands to obtain positive information, and it is impossible to optimize the brand according to user preferences. important positive hotspots

Images