Abstract
With the popularization of Internet of Things (IOT) technology, a large number of multi-source heterogeneous data are constantly generated and collected by cloud platforms, which indicates that the problem of large data in IOT has become increasingly prominent, especially for massive tags and information in IOT which is urgent to use appropriate data mining algorithms to mine the value of these data. A collaborative filtering recommendation algorithm based on multi-information source fusion (CFR-MIF) is proposed where a feature vector and time weight function are introduced to improve the accuracy of top-N recommendation. It can conveniently and effectively process the IoT data, and furthermore integrate, manage and store the massive data collected from different industries and data formats. Besides, It also provides data mining services in the whole IoT realizing prediction and decision-making, which reverses control these sensor networks, so as to control the movement and development process of objective in the Internet of Things. The experimental results based on DeviceLens 1M data set show that the proposed algorithm greatly improves the accuracy of recommendation results, recall rate and F1 value compared with other advanced algorithms.
Highlights
With the development of Internet of Things technology, the number of devices on IOT is growing exponentially
THE PROPOSED CFR-MIF ALGORITHM The steps of collaborative filtering recommendation algorithm for heterogeneous data mining in the Internet of things are as follows: step 1: build a user preference model, and convert the explicit score of the target customer into the implicit score as far as possible; Step 2: calculate the degree of asymmetric influence between users and eliminate the interference of special data as much as possible; Step 3: build the time weight function to obtain the preference degree of target users to the project at different moments; In step 4, target customers are provided with the items with the highest preference scores obtained
Based on the collaborative filtering recommendation algorithm of multi-information source fusion, we introduce a feature vector and time weight function to improve the accuracy of top-N recommendation, and we experimentalized on DeviceLens 1M data set
Summary
With the development of Internet of Things technology, the number of devices on IOT is growing exponentially. Zhang et al [21] divided the user’s historical score into several periods, analyzed the user’s interest distribution in each period, and set a time window to find the user’s recent interest These above algorithms improve the recommendation accuracy to a certain extent after the integration of time factors. They do not dig deeply into the user-item scoring matrix and do not fully consider the user’s scoring characteristics. Aims at the data sparseness problem, fully mines the score matrix information, uses the user preference model, considers the asymmetrical influence degree between users, and constructs a time weight function, a collaborative filtering recommendation algorithm based on multiple information sources fusion (CFR-MIF) is proposed in this paper as shown in fig 3. The experimental results on the DeviceLens1M datasets show that, the proposed algorithm has greatly improved the accuracy, recall rate and F1 value of the recommendation results
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 callDisclaimer: 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.
