CN104363569A - Method for recommending optimal contact ways to mobile subscriber based on context awareness - Google Patents

Abstract

The invention discloses a method for recommending an optimal contact mode to a mobile user based on situation awareness. The present invention can identify the current situation of the receiving party by using the sensor data and calendar data of the smart phone, and then accurately recommend suitable contact methods for the calling party in a specific situation, and can sort the recommended contact methods, This facilitates the use of the caller. The present invention not only considers the sensor data of the smart phone of the receiving party, but also combines the calendar data, the data will be more comprehensive and specific, and the result of situation perception is more accurate; it can not only consider the rules of the user's usage habits, but also consider the calendar information. The content is mined to recommend contact information and improve the accuracy of the recommendation; it can bind more user social data for expansion and improve the accuracy of the recommendation.

Description

A method for recommending optimal contact methods to mobile users based on context awareness

technical field

The invention relates to a contact method of a smart phone, in particular to a method for recommending an optimal contact method to a mobile user based on situation awareness.

Background technique

A smart phone refers to a personal computer, which has an independent operating system and an independent operating space. Users can install programs provided by third-party service providers such as software, games, and navigation, and can realize wireless network access through a mobile communication network. The general term for such a type of mobile phone that has been imported. Smartphones have become increasingly popular due to their greater computing power and network connectivity than traditional feature phones. Smartphones such as iPhone, Android, and Windows Phone have become more powerful in terms of computing and storage, so they can provide more functions than calling and texting. As of June 18, 2013, smartphone sales accounted for 90% of global mobile phone sales. People can connect in more ways on smartphones.

Meanwhile, social networks such as WeChat, QQ, Facebook, and Weibo have attracted billions of users. Social networks provide a good communication platform and provide convenient and fast instant messaging services. On smartphones, there are not only contact methods such as phone calls and text messages, but instant messaging software such as WeChat and QQ are also commonly used contact methods. With the diversification of the contact methods of mobile phone users, how to avoid disturbing the receiving party and obtain an early reply from the receiving party has become a concern of the calling party. Since the calling party does not know the situation of the receiving party, when the calling party wishes to contact the receiving party, the selected contact method may disturb the receiving party, or may not be able to get an immediate reply from the receiving party, thus failing to achieve an effective contact effect . In the case of a large number of contact methods, how to recommend the most appropriate contact method for a specific user's current situation has become a research problem.

Existing applications are capable of recommending contacts based on the usage data of smartphones. In this kind of recommendation, the user's use history stored in the smart phone is mainly used to recommend suitable contacts in the mobile phone address book. For example, Minetal implements the recommendation of contacts on smartphones. The recommendation system can use the historical records of smartphone users to recommend suitable contacts. In addition, the application also combines some sensor data of smartphones. The method of using the historical records of smartphone usage to recommend the receiving party's contacts, although it can recommend suitable contacts, but it is mainly aimed at recommending suitable people to contact rather than recommending to the caller when a specific person needs to be contacted Appropriate contact details. Therefore, the problem of recommending the optimal contact method for the user's current situation cannot be solved.

Another use of sensor data from smartphone users is for activity context recognition. Incremental mining is performed using sensor data from smartphones to identify the current context of the user. For example, the user's moving speed can be identified through sensor data such as speed, and then the possibility that the user is currently running can be inferred. MARS, for example, is a system that uses sensor data from smartphones to classify user activity scenarios. It can take advantage of the ever-increasing sensor perception capabilities of smartphones to obtain information about the current situation and can perform personalized situational awareness based on the user's usage history on the device. The approach of activity context recognition from smartphone sensor data has some drawbacks. Only the sensor data of the mobile phone itself (including GPS and software running status) can be recognized, and the result is not accurate enough and the accuracy is not high. First, MARS does not consider the user's calendar information, which can help determine whether the user is in a meeting or doing other things that are not suitable for being disturbed. In addition, the mobile phone data considered by MARS is not comprehensive enough, and data such as usage habits of related applications are not considered.

Contents of the invention

Aiming at the situation in the prior art that the smart phone does not consider the user's calendar information when recommending contact methods, and the mobile phone data is not comprehensive enough, the present invention provides a method for recommending optimal contact methods to mobile users based on context awareness, through Obtain the calendar data of the receiving party and the sensor data of the smart phone, and use a rule-based method to conduct context-aware analysis of the current state of the receiving party, so as to recommend the most suitable contact method in the current situation to the calling party, and at the same time combine various contacts of the receiving party According to the historical records of the method used, the recommended contact methods are sorted according to the specific sorting algorithm designed; it is very convenient to bind more social data of the receiving party to further improve the accuracy of the recommendation.

The purpose of the present invention is to provide a method for recommending an optimal contact method to a mobile user based on situation awareness.

The method for recommending an optimal contact method to a mobile user based on context awareness of the present invention comprises the following steps:

1) Establish appropriate contact methods in specific situations;

2) The mobile phone user installs the application software APP, the receiving party binds the calendar account, and adds the calling party as a friend;

3) The server regularly collects the sensor data and calendar data of the smart phones of relevant friends, saves them for processing, and uses semantic analysis and association rules to identify the current situation of the receiving party;

4) After the server recognizes the situation of the receiving party, it judges to obtain the appropriate contact method of the receiving party, and if there are multiple suitable contact methods, sorts the multiple suitable contact methods;

5) The caller sends a request to the server, and the server recommends a suitable contact method of the receiving party to the caller. If there are multiple suitable contact methods, the server recommends the sorted contact methods to the caller.

Wherein, in step 1), the contact method is divided into two types, one is the contact method of telephone voice, and the other is the contact method of text message. Telephone voice is divided into two contact methods: phone call and WeChat voice; text message is further divided into SMS and instant messaging software. Instant messaging software includes WeChat, QQ, Renren and Weibo. Compared with the contact method of text message, the voice of the phone usually has a longer ringtone and vibration, which will disturb others in certain situations, but the voice of the phone is a faster and more direct way. Therefore, if it is possible to perceive the current situation of the receiving party and judge whether the telephone voice contact method is appropriate according to the relevant situation state, it will be possible to reduce the interruption of the telephone voice contact method to the receiving party in non-emergency situations. There are short messages and contact methods of various instant messaging software in the contact information of text messages. By perceiving the usage habits of the receiving party, according to the current state of the receiving party, it recommends to the calling party a suitable text message contact method that can get a reply as soon as possible, so that Greatly reduce the user's communication cost and improve the efficiency of user contact.

(1) Analysis of Applicable Scenarios under Telephone Voice Contact

Telephone voice is the most direct method of contact, but at the same time it is also the most intrusive method for the recipient under certain circumstances. The interruption of telephone voice is mainly reflected in two aspects: on the one hand, the phone is always accompanied by long-time reminders and rings; on the other hand, the phone is a synchronous communication method that requires the receiver to respond and reply immediately The receiving party needs to stop the ongoing activity.

The situation where the telephone voice contact method is inappropriate includes the following three situations: the first situation is that the receiving party does not have a carrier signal, and the phone cannot be dialed; the second situation is that it is inconvenient to answer the call when the receiving party is in a specific environment, Such as driving, meeting or sleeping; the third situation is that the receiving party is in some noisy environment where the quality of the call cannot be guaranteed. For the three scenarios mentioned above, the telephone voice contact method is not recommended.

When deciding whether it is suitable to recommend telephone voice contact methods, the scenarios that need to be identified include the following seven:

a) Meetings/periodic activities: When the recipient is in a meeting or performing periodic activities set through the calendar, there is a high possibility of being interrupted by the phone call. Therefore, when the contact is in a meeting or performing periodic activities, it will not Recommend the contact method of telephone voice. Periodic activities refer to the periodic activities set on the calendar such as classes or regular regular meetings;

b) Driving: When the receiving party is driving, in order to comply with legal regulations and safety considerations, the telephone voice contact method will not be recommended;

c) Sleeping: If the receiving party is sleeping or taking a lunch break, the telephone voice contact method will disturb the receiving party, so in this case, the telephone voice contact method is not recommended;

d) Carrier signal: If there is no carrier signal, the phone cannot be used normally, and the telephone voice contact method will not be recommended;

e) Temporal locality: Temporal locality refers to the phone usage of the receiving party in the recent period. If the receiving party has just rejected a call recently, the possibility of being able to answer the call is very small, so telephone voice is not recommended In the same way, if you have just answered a call, it is very likely that you can answer the call, so the contact method of telephone voice is recommended;

f) Microphone usage: If the microphone is in normal use at the current situation, it means that the contact person can receive the sound signal at present. In this case, the contact method of telephone voice will be directly recommended;

g) Sound: If the mobile phone sensor finds that it is currently in a very noisy environment, the quality of the call will be affected, and the phone voice contact method is not recommended.

When the server perceives that the receiving party satisfies multiple different scenarios at the same time, the server sets a corresponding priority for the condition of each scenario. As a general rule, temporal locality and microphone usage have the highest priority. If one of the conditions is met, the phone will be recommended directly without considering other factors; when multiple scenarios are met at the same time, the order of priority is carrier signal, time locality, microphone usage, and sound, and finally In meetings, driving, or sleeping. For example, although the current ambient sound is noisy, if the contact person answered the call 1 minute ago, the server will recommend the contact method of telephone voice because the priority of time locality is higher than that of sound. Perceived contexts such as meeting, driving, or sleeping have lower priority, since all of these contexts would result in a phone not being recommended. If neither the general rules are satisfied nor the relevant context is perceived, the contact method of telephone voice is recommended by default.

(2) Analysis of applicable scenarios under the contact method of text message

Text messaging used to be the most commonly used method of text messaging on mobile phones. However, with the development of smart phones, people use more instant messaging software to send text messages, because instant messaging software is cheaper and more convenient than text messages. The server selects the most popular instant messaging software as the contact method for text messages, such as WeChat and QQ.

Since the contact method of SMS mainly uses the carrier signal of the mobile phone, and the instant messaging software such as WeChat and QQ mainly uses the wireless signal and the GPRS data flow signal, therefore, when it is suitable to recommend the contact method of text message, the scenarios that need to be identified include the following Three kinds:

a) The carrier signal is unblocked, and the wireless signal is unblocked: In this scenario, both SMS and instant messaging software are available, but because the contact method of instant messaging is cheaper, the server will give priority to recommending instant messaging software;

b) The carrier signal is smooth, but the wireless signal is not smooth: In this situation, using instant messaging software such as WeChat will also generate charges. Using instant messaging software under GPRS signal conditions cannot guarantee timely delivery of text messages. Therefore, the server recommends first SMS contact information;

c) The carrier signal is not smooth, but the wireless signal is smooth: In this situation, the text contact method of the SMS cannot be used normally, but the instant messaging software can guarantee normal use. Therefore, the server only recommends the instant messaging software.

Above, based on the analysis and introduction of appropriate contact methods in specific scenarios, the specific scenarios that need to be identified are determined. The following describes how to identify the receiving party through the analysis and mining of sensor data on smartphones and user calendar data. Whether you are currently in a specific context.

In step 3), in order to accurately perceive the situation of the receiving party, the present invention mainly includes the following five factors when identifying the situational awareness of the receiving party:

i. Receiving party information: This information can be obtained from the history log of the receiving party's smartphone;

ii. Receiving party's task schedule: These information can be calculated from the calendar of the receiving party's smartphone;

iii. Geographic location: obtained through the sensor of the receiving party's smartphone;

iv. The running status of instant messaging software (WeChat, QQ, Renren and Weibo) on smartphones;

v. The physical state of the user's environment: A large amount of information on the state of the environment can be obtained from the sensor of the smart phone of the receiving party.

The data collected by the server includes calendar data and sensor data from smartphones. Calendar data includes: text information set by users, such as periodic activities such as meetings or classes; duration of events defined by users; and reminder cycles set by users, etc. The sensor data of the smart phone is the relevant data on the smart phone of the receiving party, including wireless signal, carrier signal, last use time of the mobile phone, alarm clock information on the mobile phone, microphone usage, last call time, last missed call time, real-time Communication software running status, geographic location, speed and acceleration, etc. Through the analysis and mining of these data, more relevant information can be obtained.

Through these data, the server can identify the current situation of the user according to the application scenarios of the two different contact methods. The processing process of the present invention mainly includes the perception of applicable scenarios in the telephone voice contact mode and the perception of applicable scenarios in the text message contact mode.

Scenarios related to telephone voice contact mainly include meeting, driving, sleeping, whether there is a carrier signal, and noisy voices.

The present invention identifies the situation where the receiving party is located through semantic analysis and association rules. Semantic analysis is a scenario in which the server identifies a meeting or recurring event based on the calendar data in the receiver's smartphone. The association rule is that the server establishes a rule based on the sensor data of the receiving party's smartphone and a specific situation, so as to identify the situation of sleeping or driving. Recognition of the carrier signal, temporal locality, microphone usage, and context of the sound can be directly inferred from sensor data from the smartphone.

The context related to the text message contact is mainly related to the carrier signal and the wireless signal, and the sensor data of the smart phone acquired by the server includes the data of the carrier signal and the wireless signal. The values of the carrier signal and wireless signal indicate the strength of the connection signal. In order to ensure the smooth connection of SMS and instant messaging (WeChat, QQ, Renren and Weibo), the signal is considered to be smooth when the value of the carrier signal is greater than a certain threshold. The wireless signal also has a corresponding threshold to ensure smooth signal.

The above processing flow mainly explains how to analyze the sensor data on the receiving party's smart phone for context-aware analysis, identify the specific situation the user is in, and recommend suitable contacts to the user according to the rules of suitable contact methods in specific situations. Way. However, there may be multiple recommended contact methods that are suitable for a particular situation. Therefore, it is necessary to set a set of algorithms for sorting the recommended contact methods according to the user's history of using different contact methods.

In step 4), since the telephone voice contact method is often suitable for more urgent situations, if the receiving party’s current situation is suitable for the telephone voice contact method, the telephone voice contact method has the highest value compared with other contact methods. priority. When the receiving party's situation is suitable for both telephone voice contact and text message contact, the recommended order is that telephone voice is higher than text message contact.

Among the recommended contact methods of text messages, when both SMS and instant messaging software are recommended, since the contact method of SMS will generate a certain fee, the contact method of SMS has the lowest priority compared to other contact methods class. When the receiving party is in a situation where both text messages and instant messaging software are suitable in the contact information of text messages, the recommended order is that instant messaging software is higher than text messages.

In the recommended contact method of text messages, different users have different usage habits for different instant messaging software. Therefore, if multiple suitable instant messaging methods are recommended at the same time, the server needs to preferentially recommend the instant messaging software that the user prefers to the user. Therefore, the server will collect the historical record data of users using the instant messaging software, and by calculating their frequency of use, sort the recommended instant messaging tools.

When designing the sorting algorithm of the instant messaging software, two factors are mainly considered: the time when the first application was last used; and the frequency of the second application. The present invention does not consider the factor of whether the application is running, because on the smart phone, if the user does not actively stop the application, the application will always run in the background.

The server sorts the recommended instant messaging software according to the usage habits of the receiving party. The instant messaging software includes WeChat, QQ, Renren and Weibo. The sorting algorithm for the instant messaging software in the recommended text message includes the following steps:

A. Obtain the time when the receiving party's smart phone used the instant messaging software last time. If the instant messaging software was used in the latest time interval, the instant messaging software with the latest use time will be ranked first, and the sorting ends. If the instant messaging software is used in the latest time interval If no instant messaging software is used during the time interval, go to step B;

B. Obtain the frequency of using the instant messaging software by the receiving party's smart phone, and sort the instant messaging software according to the frequency of use.

Advantages of the present invention:

(1) Utilizing the recommended contact method of the present invention can greatly facilitate the use of the calling party. By using the sensor data and calendar data of the smart phone, the current situation of the receiving party can be identified, and then the calling party can be contacted in a specific situation. Accurately recommend suitable contact methods, and sort the recommended contact methods, so as to facilitate the caller's use;

(2) The present invention has considered the usage situation data of the smart phone, not only considered the sensor data of the smart phone, but also considered the historical log data of the mobile phone use, and also combined the calendar data of the receiving party, so the data is more comprehensive and specific, and the situation awareness The results are more accurate;

(3) The recommendation method used in the present invention combines rule-based and content-based hybrid recommendation methods, which can not only consider the rules in the user's usage habits but also consider the content in the calendar information to mine, thereby recommending contact methods and improving the recommendation. accuracy;

(4) The present invention can bind more user social data for expansion and improve the accuracy of recommendation.

Description of drawings

FIG. 1 is a flow chart of the method for recommending an optimal contact method to a mobile user based on context awareness in the present invention.

Detailed ways

The present invention will be further described through the embodiments below in conjunction with the accompanying drawings.

The method for recommending an optimal contact method to a mobile user based on context awareness of the present invention includes the following steps, as shown in Figure 1:

1) Establish appropriate contact methods in specific situations.

Contact methods are divided into two types, one is the contact method of telephone voice, and the other is the contact method of text message; telephone voice is divided into two contact methods: phone call and WeChat voice; text message is further divided into SMS and instant messaging software Two contact methods. The seven scenarios for which it is appropriate to recommend phone voice contact include: meeting/recurring activity, driving, sleeping, carrier signal, temporal locality, microphone usage, and sound. When the receiving party does not have a carrier signal, or it is inconvenient to answer the call when the receiving party is in a specific environment, or the receiving party is in some noisy environment where the quality of the call cannot be guaranteed, the telephone voice contact method is not recommended. When multiple scenarios are satisfied at the same time, the order of priority is carrier signal, temporal locality, microphone usage and sound, and finally meeting, sleeping and driving. The three scenarios of whether it is suitable to recommend the contact method of the text message include: the carrier signal is smooth and the wireless signal is smooth, the carrier signal is smooth and the wireless signal is not smooth, and the carrier signal is not smooth and the wireless signal is smooth.

2) The mobile phone user installs the application software APP, the caller binds the calendar account, and adds the caller as a friend.

3) The server regularly collects sensor data and calendar data for relevant friends, and saves them for processing, so as to perceive the situation of the receiving party.

The server can recognize the current situation of the receiving party according to the applicable scenarios of the two different contact methods. The processing process of the present invention mainly includes the perception of applicable scenarios in the telephone voice contact mode and the perception of applicable scenarios in the text message contact mode.

(1) Perception of the applicable situation under the contact method of telephone voice

Scenarios related to telephone voice contact mainly include meeting, driving, sleeping, whether there is a carrier signal, and noisy voices. The following takes meeting, driving and sleeping as examples to introduce the specific implementation methods of the corresponding situational awareness in detail.

i. Meeting/class: The server will obtain the calendar data bound by the recipient, and through semantic analysis, it can identify whether it is currently in a meeting. Through the same processing method, it can also identify whether it is currently in class. In addition, due to the ongoing cycle Therefore, we can identify whether the user is doing periodic work through the user's calendar data.

ii. Driving: It is difficult to identify the driving situation, because many factors need to be considered, and through the association rules, it is perceived whether it is in the driving situation. First, the speed should be relatively fast, around 40-100km/h; in addition, the current location should be on the road rather than indoors. The above two factors cannot be inferred to be driving, because the passengers sitting in the car also meet the above conditions. Therefore, a third factor needs to be considered, if the receiving party has not used the mobile phone for a long time, it is likely to be driving at this time. When the above three conditions are met, the server will consider that the receiving party is currently driving. Therefore, it is not suitable for telephone voice, and the server does not recommend telephone voice contact.

iii. Sleep: There is a potential association between the alarm clock on the receiving party's smartphone and sleep. Through the association rules, it is perceived whether it is in a sleeping situation. If the receiving party sets an alarm clock for getting up on the mobile phone, then within a certain period of time before the set alarm clock, for example, 20 minutes, the telephone voice contact method should not be recommended. In addition, the periodic alarm clock in the morning or at noon is likely to be an alarm clock set to wake up, and it can be perceived that the user is sleeping according to the above conditions.

(2) Perception of Applicable Scenarios under Text Message Contact Mode

The context related to the text message contact is mainly related to the carrier signal and the wireless signal, and the sensor data of the smart phone acquired by the server includes the data of the carrier signal and the wireless signal. The values of the carrier signal and wireless signal indicate the strength of the connection signal. In order to ensure the smooth connection of SMS and instant messaging, the signal is considered to be smooth when the value of the carrier signal is greater than a certain threshold, and the wireless signal also has a corresponding threshold to ensure smooth signal. Assume that the threshold of the carrier signal is K ₁ , and the threshold of the wireless signal is K _2. When the carrier signal is greater than K ₁ , it is considered that the receiver’s carrier signal is unblocked; when the wireless signal is greater than K ₂ , it is considered that the receiver is in the situation of the wireless signal. scene.

4) Judging to obtain a suitable contact method of the receiving party, and sorting the multiple contact methods if there are multiple suitable contact methods. Phone voice contacts have the highest priority over SMS text contacts.

Among the recommended contact methods of text messages, when SMS and other instant messaging software are recommended, since the contact method of SMS will generate a certain fee, the contact method of SMS has the lowest cost compared with other contact methods. priority.

In the recommended contact method of text messages, different users have different usage habits for different instant messaging software. Therefore, if multiple suitable instant messaging methods are recommended at the same time, the server sorts the contact methods of the recommended text messages according to the usage habits of the receiving party.

Take QQ and WeChat as examples below, assuming that T _w and T _q are the latest use times of WeChat and QQ respectively, and F _w and F _q are the frequency of use of WeChat and QQ respectively. Set a constant K to represent the latest time interval, and the current time is t. The sorting algorithm is as follows:

A. Obtain the time when the receiving party’s smartphone uses the instant messaging software. If t>T _w >T _q >tK, or t>T _w >tK>T _q , WeChat ranks before QQ. If t>T _q > T _w >tK, or t>T _q >tK>T _w , then QQ is ranked before WeChat, if T _w <tK and T _q <tK, go to step B;

B. Obtain the frequency of instant messaging software used by the receiving party's smartphone. If F _w > F _q , then WeChat ranks before QQ; otherwise, QQ ranks before WeChat.

5) The caller sends a request to the server, and the server recommends the appropriate contact method of the receiving party to the caller. If there are multiple suitable contact methods, the server recommends the sorted contact methods to the caller; , but also text messages, then the priority is to recommend phone voice, and when there are multiple suitable contact methods for text messages, sort the recommended contact methods for text messages to the calling party according to the habits of the receiving party.

Finally, it should be noted that the purpose of publishing the implementation is to help further understand the present invention, but those skilled in the art can understand that various replacements and modifications can be made without departing from the spirit and scope of the present invention and the appended claims. It is possible. Therefore, the present invention should not be limited to the content disclosed in the embodiments, and the protection scope of the present invention is subject to the scope defined in the claims.

Claims (9)

1., based on a method of recommending optimum contact method to mobile subscriber for context aware, it is characterized in that, said method comprising the steps of:

1) contact method suitable under setting up certain scenarios;

2) cellphone subscriber installs application software APP, the side's of answering binding calendar account, and calling party is added to good friend;

3) sensing data of the smart mobile phone of server taken at regular intervals associated buddy and calendar data, and preserve and process, utilize semantic analysis and correlation rule to identify the current residing sight in the side of answering;

4) server is after identifying the sight residing for the side of answering, and carries out judging to obtain the suitable contact method in the side of answering, and if when there is multiple suitable contact method, sorts to multiple suitable contact method;

5) call direction server sends request, and contact method suitable for the side of answering is recommended calling party by server, if when there is multiple suitable contact method, the multiple contact method after sequence is recommended calling party by server.

2. the method for claim 1, is characterized in that, in step 1) in, described contact method is divided into two kinds, and a kind of is the contact method of call voice, and another kind is the contact method of text message; Text message is divided into again the contact method of note and MSN.

3. method as claimed in claim 2, is characterized in that, when whether being applicable to the contact method recommending call voice, needs the sight identified to comprise following seven kinds:

A) meeting/cyclic activity: when answer upright in session or carry out the cyclic activity arranged by calendar time, will the contact method of call voice do not recommended;

B) driving: upright when driving when answering, will the contact method of call voice do not recommended;

C) sleep: if answer upright when sleep or lunch break, will the contact method of call voice do not recommended;

D) carrier signal: if when not having carrier signal, the contact method of call voice can not be recommended;

E) temporal locality: temporal locality refers to the side's of answering situation that phone uses within nearest a period of time, if the phone of the side's of answering just rejection recently, does not then recommend the contact method of call voice; In like manner, if just answered a phone, then the contact method of call voice has been recommended;

F) microphone service condition: if microphone is normal use under present case, will the contact method of call voice be recommended;

G) sound: if mobile phone sensor find current be in very noisy environment under, then do not recommend the contact method of call voice.

4. method as claimed in claim 2, is characterized in that, when whether being applicable to the contact method recommending text message, needs the sight identified to comprise following three kinds:

A) carrier signal is unimpeded, and wireless signal is unimpeded: in such a scenario, and server will recommend note and MSN, and

Before the order of MSN is come note;

B) carrier signal is unimpeded, and wireless signal is not smooth: in such a scenario, and server recommends note and MSN, and before note is come MSN;

C) carrier signal is not smooth, and wireless signal is unimpeded: in such a scenario, and server only recommends the MSNs such as micro-letter and QQ.

5. the method for claim 1, is characterized in that, in step 3) in, semantic analysis be server according to the calendar data in the smart mobile phone of the side of answering, thus identify meeting or the sight of cyclic activity; Correlation rule is that server sets up rule according to the sensing data of the smart mobile phone of the side of answering and specific sight, thus the sight identifying sleep or drive.

6. method as claimed in claim 2, is characterized in that, in step 4) in, when under the sight residing for the side of answering, both be applicable to the contact method of call voice, be applicable to again the contact method of text SMS, then the order of recommending is the contact method of call voice higher than text SMS.

7. method as claimed in claim 2, is characterized in that, in step 4) in, when under the sight residing for the side of answering, even if the note in the contact method of text message is all applicable with bitcom, the order of so recommending is that MSN is higher than note.

8. method as claimed in claim 2, is characterized in that, in step 4) in, server sorts to the MSN recommended according to the use habit of the side of answering, and comprises the following steps:

A. the smart mobile phone obtaining the side of answering used the time of MSN last time, if employ MSN in the nearest time interval, then come according to the MSN that service time is nearest, sequence terminates, if do not use MSN in the nearest time interval, then enter step B;

B. the smart mobile phone obtaining the side of answering uses the frequency of MSN, sorts successively according to the height of frequency of utilization to MSN.

9. method as claimed in claim 4, it is characterized in that, the threshold value of carrier signal is K ₁, the threshold value of wireless signal is K ₂, when carrier signal is greater than K ₁, then the sight that the side's of answering carrier signal is unimpeded is thought; When wireless signal is greater than K ₂time, then think that the side of answering is in the unimpeded sight of wireless signal.