Improved naive Bayesian algorithm user behavior identification method based on mobile phone sensor

Abstract

The invention provides an improved naive Bayesian algorithm user behavior identification method based on a mobile phone sensor, an improved naive Bayesian machine learning model is trained by using original data acquired by the mobile phone sensor, then classification and identification are performed on the data, and the overall process comprises the following steps: S1, data acquisition and processing; s2, feature extraction; s3, classification and identification. The method comprises the following steps: firstly, acquiring data information under different behaviors by utilizing a sensor integrated by a smart phone and preprocessing the data information; extracting characteristic attributes such as standard deviation, mean value, wave crest and wave trough, wave crest interval, correlation coefficient and the like from single-axis data of the sensor, and extracting characteristic attributes such as mean value ratio and absolute difference mean value from data between two axes; secondly, utilizing an improved naive Bayesian algorithm, wherein attribute weighting and instance weighting are combined in the algorithm, the core is that attribute weight is brought into a naive Bayesianclassification formula; and utilizing instance weighting training data to estimate prior probability and conditional probability; and finally, obtaining the prior probability of each classification according to the training set, then obtaining the posterior probability of the unknown class sample, and obtaining the classification of the unknown class sample by comparing the probabilities to realize the classification of the behaviors.

Description

Technical field: [0001] The invention relates to a user behavior recognition method, in particular to an improved naive Bayesian behavior recognition method based on mobile phone sensors. [0002] technical background: [0003] User behavior recognition technology is the process of obtaining user behavior patterns by analyzing the user's external behavior. It has broad application prospects in artificial intelligence, pattern recognition and other fields. Since sensor data can reflect different characteristics of human behavior, there are currently a large number of researches on wearable sensors. For example, a combination of three-axis acceleration sensor and gyroscope is used to construct an activity sensing device for the elderly. Two acceleration sensors are worn on the right arm. Before and After Solving the Problem of Upper Limb Action Recognition in Interactive Games. However, researchers need to set up special motion sensors in different body parts, such as arms, wa...

AI Technical Summary

Problems solved by technology

However, researchers need to set up special motion sensors in different body parts, such as arms, waist, thighs, wrists, ankles, etc., to capture behavioral data. These sensors affect the wearer's daily work and cannot be used for activity monitoring or behavior prediction. Provide long-term and effective solutions, while wearable devices have problems such as high market price and inconvenient daily wear. Then the research hotspot is transferred to smart phone sensors

Images