Article comment processing method and system, electronic equipment and readable storage medium
A processing method and technology of a processing system, applied in the field of item review processing method, system, electronic equipment and readable storage media, can solve the problems of low labeling efficiency and low automation of item review processing, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0058] A method for handling item reviews such as figure 1 As shown, the processing method includes:
[0059] Step 10, constructing an attribute list of an item category;
[0060] The attribute list includes the attribute categories of all items under the item category and the corresponding attributes of each attribute category; according to the classification of an item, it can be divided into several attribute categories, and each attribute category is an item. A general term for various specification parameters, a certain attribute under a certain specification parameter, such as a display parameter is an attribute category, including brightness, screen ratio, refresh rate, scanning mode, response time, screen resolution, 3D display and other attributes.
[0061]Step 20, crawling all comment data of a target item under the item category within a preset time period;
[0062] It should be noted that data can be crawled by SKU for a single item, or by SPU for items of the sa...
Embodiment 2
[0078] The processing method of article comment in this embodiment is further improved on the basis of embodiment 1, such as figure 2 As shown, the step 60 of calculating the similarity between each attribute and the participle in the participle set specifically includes:
[0079] Step 6011, for each word-segment set, respectively convert the attribute and the word-segment in the word-segment set according to the phonetic-shaped coding rules, and generate the corresponding phonetic-shaped code;
[0080] Step 6012: Based on the edit distance calculation model, calculate the degree of similarity between the phonetic-shaped coding of the attribute and the phonetic-shaped coding of the participle in the participle set to obtain a similarity.
[0081] Each attribute can correspond to different expressions, such as screen resolution, expression A: ultra-high-definition 4K or full HD or high-definition, expression B: the video looks very clear, or more colloquial, watching movies is...
Embodiment 3
[0085] The processing method of article comment in this embodiment is further improved on the basis of embodiment 1, such as image 3 As shown, step 60 specifically includes:
[0086] Step 6021, split the word segmentation set into multiple word segmentation sub-sets;
[0087] Step 6022, multithreading simultaneously calculates the similarity of multiple word segmentation sub-sets;
[0088] Step 6023, merging the calculation results of each word segmentation sub-set;
[0089] Step 6024, obtain the frequency according to the combined calculation result.
[0090] In order to adapt to the data processing of millions and tens of millions, the word segmentation set can be split and processed according to the time dimension. Specifically, the set size can be manually set and flexibly adjusted. After processing separately, the final list can be obtained from the set statistics.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


