Activity recommendation method and system based on intelligent door lock opening rule
Technical Field
The invention relates to the field of data processing, in particular to an activity recommendation method and system based on an intelligent door lock door opening rule.
Background
The long rental apartment is a new type of house renting, including apartments, eggshell apartments and the like, and is generally in a way of renting together, which is a house renting way more meeting the requirements of tenants. A big house is independently divided into a plurality of small rooms, each tenant has an independent room, the door of the independent room is an inner door, and the living room can be used in common for communication or leisure of the tenants. The whole large house also has a common outer door.
At present, most of leased rooms or long-rented apartments are managed by intelligent door locks. The password is managed by the house renting platform, and the tenant can enter and exit the room inner door and the room outer door through the password. The intelligent door lock can record the time condition of opening and closing the door and the like.
On the other hand, a rental platform (e.g., a liberty platform) recommends platform-related activities to the tenant through the app, including a wide variety of activities such as reading books, traveling, playing, gathering, and the like. The activity recommendations are not adaptively adjusted according to the real behavior rules of the tenant in the rental house. Therefore, the most accurate activity cannot be recommended to the tenant, the enthusiasm of the tenant for using platform software is reduced, and the recommendation effect is poor.
In order to achieve accurate recommendation, some software can snoop the privacy of the tenant, and a camera is used for monitoring the behavior or the position of the tenant, so that the effect is better, but the privacy of the tenant is greatly invaded.
Disclosure of Invention
The invention provides an activity recommendation method and system based on an intelligent door lock door opening rule, which are used for recommending activities according to the similarity of tenants.
The invention provides an activity recommendation method based on an intelligent door lock door opening rule, which mainly comprises the following steps:
acquiring the door opening and closing time of an inner door and an outer door provided with an intelligent door lock; the inner door and the outer door are that a large house is independently divided into a plurality of small rooms by a long renting apartment, each tenant has an independent room, the door of the independent room is the inner door, a living room can be shared for communication or leisure of the tenants, and the whole large house is also provided with a shared door which is the outer door;
analyzing the activity positions of different tenants according to the opening and closing conditions of the inner door and the outer door of the intelligent door lock; the activity positions comprise rooms, classrooms and outdoors;
carrying out statistical analysis on time periods of the tenants at different activity positions;
according to different time periods and different activity positions of the tenants, calculating the similarity among the tenants;
and performing activity recommendation on the tenant.
Further optionally, in the method, the analyzing the activity positions of different tenants according to the opening and closing conditions of the inside and outside doors of the intelligent door lock mainly includes:
acquiring the time length of each tenant in a room;
acquiring the time length of each tenant in the living room; the length of time the tenant is out of doors is obtained.
Further optionally, in the method as described above, the counting the length of time that the tenant is outdoors, further includes:
the adjustment and error correction of the time of going out mainly comprises,
when the first tenant leaves the inner door, no record of leaving the outer door exists; later, again from the outside door; and synchronizing the time when the first tenant leaves the external door with the time when the second tenant nearest to the time when the first tenant leaves the external door after the first tenant leaves the internal door to the first tenant, wherein the time is used as the time when the first tenant leaves the external door.
Further optionally, in the method as described above, the performing statistical analysis on the time periods of the different activity locations of the tenant mainly includes:
counting the weekend trip time length of the tenant;
and counting the time period that the tenant is used to go out.
Further optionally, in the method, the calculating the similarity between tenants according to different time periods and different activity locations of the tenants mainly includes:
and constructing a characteristic matrix according to the indoor and outdoor time lengths of the tenants and the time period data habitually going out, clustering the tenants to obtain users with larger similarity, and recommending similar activities for the users clustered in the same category.
Further optionally, in the method as described above, the recommending an activity to the tenant mainly includes:
and judging the character characteristics of the tenants according to the time when the tenants are rented in the apartment and out of the apartment, and recommending related activities based on a collaborative filtering method.
The invention provides an activity recommendation system based on an intelligent door lock door opening rule, which comprises:
the acquisition module is used for acquiring the door opening and closing time of the inner door and the outer door of the intelligent door lock;
the activity area analysis module is used for analyzing activity positions of different tenants at different moments according to the opening and closing conditions of the inner door and the outer door;
the system comprises an outgoing time period analysis module, a data processing module and a data processing module, wherein the outgoing time period analysis module is used for analyzing the outgoing time of the tenant habit;
the similarity calculation module is used for calculating behavior similarity among the tenants;
and the activity recommendation module is used for recommending activities according to the tenant similarity.
The invention avoids the method that a large amount of user privacy can be revealed through mobile phone position positioning, a camera and the like, and achieves the purposes of analyzing the user character and not monitoring the user pair too much.
The behavior characteristics of the user are analyzed by the intelligent door lock, the user does not need to actively participate in other user portrait methods, and the user can obtain data through passive entry and exit.
The method has the advantages that the obtained user data is efficient and accurate, and the user privacy is well protected.
Drawings
FIG. 1 is a flow chart of an activity recommendation method based on the door opening rule of an intelligent door lock according to the invention;
fig. 2 is a structural diagram of an embodiment of the activity recommendation system based on the door opening rule of the intelligent door lock.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings and specific embodiments.
Fig. 1 is a flowchart of an embodiment of an activity recommendation method based on an intelligent door lock door opening rule according to the present invention. As shown in fig. 1, the activity recommendation method based on the door opening rule of the intelligent door lock in the embodiment may specifically include the following steps:
a long rental apartment is formed by independently dividing a large house into a plurality of small rooms, each tenant has an independent room, the door of each independent room is an inner door, and the living room can be shared for communication or leisure of the tenants. The whole large house also has a common outer door.
And step 100, recording the door opening conditions of the inner door and the outer door of the tenant by the intelligent door lock, and recording the door opening condition of inputting the password every time. The door is divided into an outward door and an inward door. Still another type is an entry door, which opens and closes the door from the outside in. The difference is that the password is not required to be input when going out, and the password is required to be input every time when going in.
Step 101, counting the time of each tenant in the room as a first data characteristic. The recording method is that when the tenant inputs the password, the tenant enters the inner door to start to calculate the time. The time is stopped when the tenant leaves. A tenant who likes to stay in a room for a long time has a low probability of liking outdoor activities. The character may be more inward biased. It is suitable for recommending quieter activities such as reading books and watching movies in libraries.
And step 102, counting the time of each tenant in the living room as a second data characteristic. The calculation method is to count the time when the tenant leaves the internal door but does not leave the external door. When the tenant in the living room goes out of the outer door again or the tenant enters the inner door again, the time is over. The longer tenant in the living room shows that the tenant does not exclude collective life and is more willing to communicate with the roommates. But may be less likely to be too tiring for outdoor activities.
And step 103, counting the time of the tenant outdoors as a third data characteristic. And counting is started when the tenant leaves the outer door, and counting is stopped when the tenant returns from the outside and enters the outer door. The method for verifying the entrance of different tenants into the external door is to judge the difference of the entered tenants according to different tenant passwords. The behavior of the frequent out-going renters is more outward. Particularly, people often go out at weekend time, and rent guests regularly go home, and the outdoor activities are preferred.
And step 104, correcting the unreasonable statistical condition. For example, when a tenant leaves an inside door, there is no longer a record of leaving the outside door. And the situation that the door enters from the outer door after a long time. Indicating that the tenant is going out simultaneously with other tenants. Thus synchronizing the time of the other tenant's exit to the tenant. For calculating the actual time of departure of the tenant.
And 105, according to the door access record of the intelligent door lock of the long rental apartment, performing sectional recording on the weekend trip time period of the tenant as a fourth data characteristic. For example, divided into weekend trips and weekday trips, and the time of the trip of the tenant is counted. Therefore, the travel of the weekdays is regular, and the weekend travel can judge the preference of one person, and the people who like going out on the weekend prefer the going-out activities. People who like home are not as interested in the activity.
And step 106, counting favorite outgoing time periods as a fifth data characteristic. For example, each half hour is divided into a time period, and the like degree of the tenant to the event holding time period is judged according to the time period. The time period is also one of judgment bases of the similarity of the tenants. When the renting addresses are close, people with the same trip time period have higher occupation similarity. For example, the internet industry generally has a time of day nine hours in the morning, while the officer generally has a time of day nine hours in the morning. The home time period is also a method for judging the similarity of the tenants. Some people like night life, and some people do not like. The renters who like the activities of living at night are called home for a later time period. Suitable for recommending activities with a holding time at a later time.
And step 107, extracting the characteristics based on the five data characteristics to form a characteristic matrix. And clustering the out time and behavior of the tenant. And (4) gathering the tenants with similar characters and similar outgoing time rules. And recommending the tenants with similar activity behavior rules to associate and participate in the activity together.
The similarity calculation method may measure according to a similarity calculation method adopted by clustering, for example, clustering is performed by using a bitch method. The method can not define the initial category number of the users, and the users can be gathered into different categories according to the similarity degree as long as the initial similarity threshold is set. Users in the same category have certain behavior similarity. For example, tenant 2 and tenant 3 above have a higher degree of similarity.
And step 108, judging the character characteristics of the tenants according to the time when the tenants rent in the apartment and go out, and recommending the tenants with similar characteristics but with certain activities already gone to the tenants which have not participated in based on a collaborative filtering method.
The method for judging the character features is to judge according to the methods for judging the internal and external directions, the career tendency and the like in the steps 100 to 106.
The algorithm adopts a user-based collaborative filtering method, and recommends activities of other tenants which are more similar to the current tenant according to the similarity between tenants.
Fig. 2 is a structural diagram of an activity recommendation system based on an intelligent door lock door opening rule, which includes: the system comprises an acquisition module and an activity area analysis module, wherein the activity area analysis module is used for analyzing activity positions of different tenants at different moments according to opening and closing conditions of an inner door and an outer door; the system comprises an outgoing time period analysis module, a data processing module and a data processing module, wherein the outgoing time period analysis module is used for analyzing the outgoing time of the tenant habit; the similarity calculation module is used for calculating behavior similarity among the tenants; and the activity recommendation module is used for recommending activities according to the tenant similarity.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.