Abstract
The sparsity problem could result in a data-dependent reduction and we couldn't do rough set null value estimates, therefore, we need to deal with the problem of a sparse data set before performing the null value estimate and padded by introducing a collaborative filtering technology used the sparse data processing methods - project- based score prediction in the study. The method in the case of the object attribute data sparse, two objects can be based on their known attributes of computing the similarity between them, so a target object can be predicted based on the similarity between the size of the other objects to the N objects determine a neighbor collection of objects and then treat the predicted target unknown property by neighbors object contains attribute values to predict.
Highlights
Sparsity problem is one of the priorities of the recommended techniques of data tables that contain a large number of null values due to appear in the actual recommendation system often, collaborative filtering technology to achieve firstly is to deal with the sparsity of the data table, otherwise, the e-commerce the system would not be able to type of data for processing.information system containing a large number of null values for subsequent data processing has brought great difficulties and cannot generate accurate and effective decision-making rules.Collaborative filtering the processing object technology can be a two-dimensional table of data, the same, object handling in rough set theory is a two-dimensional table, can score prediction method using collaborative filtering technology the sparse information systems rough set data processing.Nearest-based collaborative filtering recommendation algorithm needs to measure the similarity between different users and select the highest number of user’s similarity to the current user as the current user's nearest neighbor set, the last collection by the recommendation algorithm based on neighbor ratings recommended produce results (Yu et al, 2001; Krasowski, 1988; Zou et al, 2001; Li, 2001; Zhang et al, 2003)
Nearest-based collaborative filtering recommendation algorithm needs to measure the similarity between different users and select the highest number of user’s similarity to the current user as the current user's nearest neighbor set, the last collection by the recommendation algorithm based on neighbor ratings recommended produce results (Yu et al, 2001; Krasowski, 1988; Zou et al, 2001; Li, 2001; Zhang et al, 2003)
Promotion rough set field to predict the value of an object's empty, you need to measure the similarity between the object and other objects and select the highest similarity with the object of several objects most collection of its neighbors, by valuation algorithm according to the neighbor set of attribute values null value of the object
Summary
Sparsity problem is one of the priorities of the recommended techniques of data tables that contain a large number of null values due to appear in the actual recommendation system often, collaborative filtering technology to achieve firstly is to deal with the sparsity of the data table, otherwise, the e-commerce the system would not be able to type of data for processing.information system containing a large number of null values for subsequent data processing has brought great difficulties and cannot generate accurate and effective decision-making rules.Collaborative filtering the processing object technology can be a two-dimensional table of data, the same, object handling in rough set theory is a two-dimensional table, can score prediction method using collaborative filtering technology the sparse information systems rough set data processing.Nearest-based collaborative filtering recommendation algorithm needs to measure the similarity between different users and select the highest number of user’s similarity to the current user as the current user's nearest neighbor set, the last collection by the recommendation algorithm based on neighbor ratings recommended produce results (Yu et al, 2001; Krasowski, 1988; Zou et al, 2001; Li, 2001; Zhang et al, 2003). Based on the above two aspects, we propose a double feature weight method, from the data set, respectively, "horizontal" and "vertical" two considerations to consider the characteristics of the property itself weights, and consider the degree of association between the object make up the case of the entropy weights law failure in small differences in property values, our similarity calculation formula (?) the following improvements: sim(i, j) =
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Research Journal of Applied Sciences, Engineering and Technology
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
