A product ranking method based on word-of-mouth product characteristics
A sorting method and product technology, applied in special data processing applications, instruments, calculations, etc., can solve problems such as inaccurate comparison of product features, inability to process, and deviations in sorting results, and achieve accurate and reliable sorting results
Active Publication Date: 2020-09-29
DALIAN UNIV OF TECH
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The method proposed in this paper also has the following deficiencies: First, the text mining methods used in this method, such as product feature extraction and sentiment classification, are all aimed at English text reviews and cannot be directly used to process Chinese text reviews; When extracting relationships, the method of this paper only extracts the comparative relationship between products at a coarse-grained level, and does not identify the comparative relationship between product features at a fine-grained level, and the established comparative relationship between product features is inaccurate
Finally, when products are sorted, the emotional score of product features does not take into account the importance of product features, which will lead to biased ranking results
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0040] The implementation of the present invention will be described in detail below, taking three mobile phone products P1, P2, and P3 as examples.
[0041] The first step is to identify product features, process all text reviews of the three mobile phone products, and identify product features. Product features are arranged in descending order according to the frequency of mentions. The top five most mentioned product features are: { Cost performance, battery life, screen, appearance, memory}, the frequency of mentioning each product feature of the three specific products is shown in Table 3.
[0042] Table 3 The frequency of mentioning each product feature of the three products
[0043]
[0044] In the second step, for the three products, respectively, identify the opinion word sets of the three products about each product feature. And through the opinion words and their emotional strength, the emotional score of each product feature is calculated according to the formul...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention belongs to the technical field of data mining and provides a product sorting method based on public praise of product characteristics. The method comprises the following steps that 1) the product characteristics are identified, all comments of candidate products are subjected to word segmentation, part-of-speech annotation and pruning, and a product characteristic set is established;2) according to the obtained product characteristic set, opinion words of the product characteristics are identified, and emotion values of the product characteristics are calculated; 3) a comparisonrelation of the product characteristics in the text comments is identified; 4) according to the emotion values of the product characteristics of the products and the number of polls obtained by comparing the product characteristics with other products, public praise values of the product characteristics are calculated, and after the public praise values of the product characteristics are obtained, the products can be sorted in a descending order on basis of the public praise of all the product characteristics. According to the product sorting method based on the public praise of the product characteristics, the public praise of the products can be sorted for different and multidimensional product characteristics, and a provided sorting result is more precise and reliable in combination with more comprehensive heterogeneous information.
Description
technical field [0001] The invention belongs to the technical field of data mining and relates to a method for quantifying word-of-mouth information of product features in text comments and sorting products. Background technique [0002] Product ranking based on word of mouth can avoid information overload and provide effective decision support for consumers and merchants on e-commerce websites, so it has become a research focus and hotspot in related fields. Online reviews are an important form of product reputation. At present, online reviews on e-commerce websites mainly include ratings (1-5 stars) and text reviews. When these websites sort the word-of-mouth of products, they generally ignore the text reviews, and only consider the ratings as the total word-of-mouth of the products to rank the products. However, the content of the text review is more detailed, including word-of-mouth information rich in product features (such as: product features of mobile phones—cost pe...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/332G06F40/284G06F40/247
CPCG06F16/3344G06F40/247G06F40/284
Inventor 杨弦杨光飞吴江宁党延忠
Owner DALIAN UNIV OF TECH