Advertisement recommendation method and device
A recommendation method and advertising technology, applied in advertising, equipment, network data retrieval, etc., can solve the problems of incomplete matching or relevance, low recommendation efficiency, and failure to consider users' own interests, etc., to achieve the effect of improving recommendation efficiency
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0056] The present invention provides an advertising recommendation method, such as Figure 1 shown, including:
[0057] Step 1: Collect the user's historical behavior record and current behavior record on the target application;
[0058] Step 2: According to the historical behavior record and the current behavior record, determine the user characteristic vector of the user;
[0059] Step 3: Collect several existing advertisements and analyze the advertising characteristic vectors of each advertisement;
[0060] Step 4: Calculate the user characteristic vector and the cosine vector value of each advertising characteristic vector;
[0061] Step 5: Filter the maximum cosine vector value from all cosine vector values and get the target ad corresponding to the maximum cosine vector value;
[0062] Step 6: Recommend the said targeted advertisement to the said user.
[0063] In this embodiment, the historical behavior record comprises: the device identification of the historical device use...
Embodiment 2
[0071] Based on Example 1, according to the historical behavior record and the current behavior record, the user characteristic vector of the user is determined, comprising:
[0072] According to the historical behavior record and the current behavior record, the user is assigned the behavior bias label and the label weight of each behavior bias label;
[0073] Based on the behavior bias label and the label weight of each behavior bias label, the user characteristic vector of the user is obtained.
[0074] In this embodiment, in the process of obtaining behavior bias tags, the historical behavior records and the current behavior records are filtered according to the tag configuration, that is, the tags can be used to distinguish the interrelated messages under the same tag type, for example, one room, two rooms, three rooms, and four rooms under the type of label, at this time, if the historical behavior records are more biased towards the search of two rooms, at this time, the one ...
Embodiment 3
[0078] Based on Example 2, on the basis of example 2, a plurality of existing advertisements are collected, and the advertising characteristic vectors of each advertisement are analyzed, comprising:
[0079] Assign ad bias labels and label weights for ads bias labels to each ad collected;
[0080] Based on the advertising bias label of the same advertisement and the label weight of the advertisement bias label, the advertising characteristic vector of the same advertisement is obtained.
[0081] In this embodiment, the advertising bias label assigned to the advertisement is marked in advance by the operator, and in the process of assigning the label weight, the weight may be set according to the popularity of the label.
[0082] For example, the labels are b and d, and the weight values are v4 and v5, respectively.
[0083] The beneficial effect of the above technical solution is: by directly assigning labels and weights to the advertisement, it is convenient to obtain the advertisi...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More