CN110705957A - Suggestion result feedback method of man-machine cooperative type weak artificial intelligence cloud system - Google Patents

Abstract

The invention proposes a method for feedback of suggestion results of a human-machine collaborative weak artificial intelligence cloud system, which belongs to the field of artificial intelligence. In the scenario of a multi-person, one-machine collaborative artificial intelligence cloud service, when the number of personal databases that have different and the same recommendation results from the professional database exceeds half of the total number of personal databases, the method of the present invention first calculates the The comprehensive weight of individual databases with different and the same recommendation results is recommended, and then it is judged whether the comprehensive weight is greater than the professional data update decision value. When the judgment result is yes, the corresponding data in the professional database is updated. Otherwise, the data of the professional database is not updated. After the above steps are implemented, the artificial intelligence algorithm is used to obtain the suggestion results based on the current professional database, and the results are fed back to the user, so as to solve the problem of human-machine collaborative weak artificial intelligence cloud system when the results of multiple personal suggestions are uneven and the same as the professional suggestion results. Suggested results feedback questions.

Description

Suggestion result feedback method for human-machine collaborative weak artificial intelligence cloud system

technical field

The invention relates to a data updating method of a database, belonging to the field of artificial intelligence.

Background technique

The existing artificial intelligence (AI) can be simply divided into strong artificial intelligence and weak artificial intelligence according to its intelligence.

Strong artificial intelligence believes that it is possible to create intelligent machines that can truly reason and solve problems. Such machines are considered to be sentient, self-aware, able to think independently and formulate optimal solutions to problems, have their own systems of values and worldviews, and have the same instincts as living creatures, such as survival and safety needs . In a sense, such machines can be seen as a new civilization.

Weak AI believes that it is impossible to create intelligent machines that can truly reason and solve problems. Such machines only appear to be intelligent, but they are not really intelligent and have no autonomous consciousness. Such machines only provide suggested results and still require humans to make the final decision.

Human-machine collaborative architecture is the development trend of existing weak artificial intelligence cloud systems. Human-machine collaborative weak artificial intelligence cloud systems take artificial intelligence algorithms as the core, rely on professional databases and personal databases, and provide users with artificial intelligence online services. The process for the human-machine collaborative weak artificial intelligence cloud system to obtain the recommended results is: the cloud host forwards the questions sent by the user to the cloud server. After receiving the user's question, the cloud server obtains professional advice results based on artificial intelligence algorithms and professional databases. At the same time, the cloud server obtains multiple personal advice results based on artificial intelligence algorithms and multiple personal databases. Therefore, the feedback of the proposed results of the human-machine collaborative weak artificial intelligence cloud system will inevitably involve the following issues:

How does the human-machine collaborative weak artificial intelligence cloud system give feedback to users when the results of multiple personal suggestions are uneven and the same as professional suggestions.

SUMMARY OF THE INVENTION

In order to solve the above problems of the human-machine collaborative weak artificial intelligence cloud system, the present invention proposes a suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system.

The proposed result feedback method of the human-machine collaborative weak artificial intelligence cloud system according to the present invention includes:

In the scenario of multi-person, one-machine collaborative artificial intelligence cloud service, the number of personal databases that use artificial intelligence algorithms and are based on one professional database and multiple personal databases to deal with the same problem, and the results of which are different from the professional database and the same as each other's suggestion results exceeds the number of personal databases. When half of the total number of databases, calculate the comprehensive weight of personal databases with different recommendations from professional databases and the same as each other;

Judging whether the comprehensive weight is greater than the predetermined professional data update decision value;

When the judgment result is yes, update the corresponding data of the professional database based on the question and the suggestion result given by the personal database which is different from the professional database suggestion result and the same as each other's suggestion result;

When the judgment result is no, the corresponding data of the professional database will not be updated;

Using artificial intelligence algorithms to obtain recommendations based on the current professional database, and feedback the recommendations to users;

The database recommendation results are the recommendation results obtained based on the database using artificial intelligence algorithms.

Preferably, the calculation method of the comprehensive weight includes:

evenly distribute the weights of the plurality of personal databases, and calculate the total weights of the personal databases that have different suggested results from the professional database and have the same suggested results from each other;

For personal databases that are different from the professional database recommendation results but the same as each other, obtain the product of the weight of each database and the established weight proportional coefficient of the database, and use the sum of all the products as the result that is different from the professional database recommendation results and mutually recommended results The total weight of the same personal database;

The average value of the two total weights obtained by implementing the above steps is taken as the comprehensive weight.

Preferably, the database weight proportional coefficient is the acceptance rate of the suggestion results previously given by the corresponding database within a predetermined period of time.

Preferably, the professional data update decision value is a ratio of the number of suggested results that are different from and identical to the professional database suggested results given by the plurality of personal databases within a predetermined period to the total number of suggested results.

Preferably, the updating of the corresponding data in the professional database includes:

Store the question and the suggestion results given by the personal database which is different from the professional database suggestion result and the same as each other's suggestion result as professional data in the professional database or replace the original corresponding data in the professional database;

Train on specialized databases with updated data.

The proposed result feedback method of the human-machine collaborative weak artificial intelligence cloud system of the present invention adopts artificial intelligence algorithms and is based on a professional database and multiple individuals in the scenario of multi-person, one-machine collaborative artificial intelligence cloud services. When the database deals with the same problem, and the number of individual databases with the same recommendation results as those of the professional database and the same recommendation results is more than half of the total number of individual databases, the comprehensive weight of the individual databases with different recommendation results from the professional database and the same recommendation results as each other is calculated first. It is then judged whether the comprehensive weight is greater than the predetermined professional data update decision value. When the judgment result is yes, the corresponding data of the professional database is updated based on the question and the suggestion result given by the personal database which is different from the professional database suggestion result and the same as each other. When the judgment result is no, the corresponding data of the professional database is not updated. When the professional database data is updated or the professional database does not need to update the data, the artificial intelligence algorithm is used to obtain the suggestion result based on the current professional database, and the suggestion result is fed back to the user, so as to solve the problem of uneven personal suggestion results and professional suggestion results. At the same time, the suggestion result feedback problem of the human-machine collaborative weak artificial intelligence cloud system.

Description of drawings

Hereinafter, based on the embodiments and with reference to the accompanying drawings, the proposed result feedback method of the human-machine collaborative weak artificial intelligence cloud system according to the present invention will be described in more detail, wherein:

Fig. 1 is the realization flow chart of the suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in the embodiment;

Fig. 2 is the realization flow chart of the calculation method of the comprehensive weight of the personal database mentioned in the embodiment that is different from the professional database suggestion result and the same as each other's suggestion result;

Fig. 3 is the realization flow chart that the corresponding data of the specialized database is updated as mentioned in the embodiment;

FIG. 4 is a system frame diagram of the human-machine collaborative weak artificial intelligence cloud system mentioned in the embodiment.

Detailed ways

The method for feeding back the suggested results of the human-machine collaborative weak artificial intelligence cloud system according to the present invention will be further described below with reference to the accompanying drawings.

Embodiment: The present embodiment will be described in detail below with reference to FIG. 1 to FIG. 4 .

The proposed result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment is suitable for the scenario of multi-person one-machine collaborative artificial intelligence cloud service, adopts artificial intelligence algorithm and is based on a professional database and multiple personal databases Cases where the number of individual databases dealing with the same issue with different and mutually identical advice results than those of professional databases exceeds half of the total number of individual databases, including:

Step S1, calculating the comprehensive weight of personal databases that are different from the professional database suggestion results and have the same suggestion results as each other;

Step S2, judging whether the comprehensive weight is greater than the predetermined professional data update decision value, when the judgment result is yes, execute S3, otherwise, execute S4;

Step S3, update the corresponding data of the professional database based on the problem and the suggested results provided by the personal databases that are different from the professional database suggestion results and are identical to each other's suggestion results;

Step S4, do not update the corresponding data of the professional database;

Step S5 , using artificial intelligence algorithms to obtain a suggestion result based on the current professional database, and feeding the suggestion result back to the user.

In this embodiment, the suggestion result of the professional database and the suggestion result given by the database both refer to the suggestion result obtained based on the database using an artificial intelligence algorithm. In this embodiment, the multi-person and one-machine in the multi-person, one-machine collaborative artificial intelligence cloud service refers to multiple personal databases and one professional database. When using artificial intelligence algorithms to deal with the same problem based on a professional database and multiple personal databases, the recommendation results of the professional database are more authoritative, and the recommendation results of the personal database are for reference only because they are based on personal experience. However, personal experience will increase over time. When faced with the same problem, the number of personal databases with different and the same recommendation results from professional databases exceeds half of the total number of personal databases, the final recommendation result of the artificial intelligence cloud service needs to be reconsidered. However, the method for feeding back the suggested results of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment is exactly applicable to the final feedback of the suggested results of the human-machine collaborative weak artificial intelligence cloud system under the above circumstances.

The suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment only considers the situation that the number of personal databases with different and the same suggestion results from professional databases exceeds half of the total number of personal databases. In cases where the number of personal databases with different database recommendation results and the same recommendation results as each other does not exceed half of the total number of personal databases, in this case, there is no need to update the corresponding professional data of the professional database.

The frame diagram of the human-machine collaborative weak artificial intelligence cloud system based on the suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment is shown in FIG. 4 .

In the suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment, the calculation method of the comprehensive weight of the personal database which is different from the professional database suggestion result and the same as each other's suggestion result includes:

Step S11, equally distribute the weights of the multiple personal databases, and calculate the total weights of the personal databases that are different from the professional database suggestion results and are the same as each other's suggestion results;

Step S12, for the personal databases that are different from the professional database suggestion results and the same as each other's suggestion results, obtain the product of the weight of each database and the predetermined weight proportional coefficient of the database, and use the sum of all products as the professional database. The total weight of individual databases that suggest the same results as each other;

In step S13, the average value of the total weight obtained by implementing step S11 and the total weight obtained by implementing step S12 is used as the total weight.

In the method for feeding back the suggestion results of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment, the database weight proportional coefficient is the acceptance rate of the suggestion results previously given by the corresponding database within a predetermined period of time.

In the suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment, the professional data update decision value is different from the professional database suggestion results given by the plurality of personal databases within a predetermined period of time and the ratio of the number of suggested results that are identical to each other to the total number of suggested results.

The proposed result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment proposes the concept of comprehensive weight. The problem of the adoption rate of the recommendation results of the personal database makes the calculated comprehensive weights of the personal databases that are different from the professional database recommendation results and the same as each other's recommendation results better reflect the recommendations of the personal databases that are different from the professional database recommendation results and the same as each other's recommendation results. The authoritativeness of the results and the need for data updates to specialized databases.

The professional data update decision value proposed by the suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment is different and identical to the professional database suggestion results given by the plurality of personal databases within a predetermined period of time The ratio of the number of suggested results to the total number of suggested results. For example, in 10 days, 10 personal databases and 1 professional database jointly dealt with 10 questions. During the 10 processing of the questions, the number of personal databases with different and the same suggested results from the professional database is 3, 4, 6, 2, 5, 3, 5, 2, 5 and 4, then the corresponding professional data update decision value is (3+4+6+2+5+3+5+2+5+4)/100 . The predetermined period here refers to a historical period of a certain length. It can be seen from this that the professional data update decision value proposed by the recommendation result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment can truly reflect the different and mutually recommended results from the professional database within a certain length of history. The same proportion of personal databases ensures the necessity of updating professional database data.

In the proposed result feedback method of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment, the updating of the corresponding data of the professional database includes:

Step S31, storing the problem and the suggestion result provided by the personal database that is different from the professional database suggestion result and identical to each other's suggestion result as professional data into the professional database or replace the original corresponding data in the professional database;

Step S32, training the professional database after the data update;

Step S33, judging whether the training of the professional database is completed, when the judgment result is yes, go to step S34, otherwise, go to step S31;

Step S34, end.

In the method for feeding back the suggested results of the human-machine collaborative weak artificial intelligence cloud system described in this embodiment, in the process of updating the corresponding data of the professional database, it is determined whether the training of the professional database is completed through step S33, and when the training is not completed, Continue to train the professional database until the training of the professional database is completed, thereby ensuring the security of the professional data update of the professional database.

Although the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely illustrative of the principles and applications of the invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and other arrangements can be devised without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that the features described in the various dependent claims and herein may be combined in different ways than are described in the original claims. It will also be appreciated that features described in connection with a single embodiment may be used in other described embodiments.

Claims (5)

1. The proposed result feedback method of the human-machine collaborative weak artificial intelligence cloud system is characterized in that, comprising: In the scenario of multi-person, one-machine collaborative artificial intelligence cloud service, the number of personal databases that use artificial intelligence algorithms and are based on one professional database and multiple personal databases to deal with the same problem, and the results of which are different from the professional database and the same as each other's suggestion results exceeds the number of personal databases. When half of the total number of databases, calculate the comprehensive weight of personal databases with different recommendations from professional databases and the same as each other; Judging whether the comprehensive weight is greater than the predetermined professional data update decision value; When the judgment result is yes, update the corresponding data of the professional database based on the question and the suggestion result given by the personal database which is different from the professional database suggestion result and the same as each other's suggestion result; When the judgment result is no, the corresponding data of the professional database will not be updated; Using artificial intelligence algorithms to obtain recommendations based on the current professional database, and feedback the recommendations to users; The database recommendation results are the recommendation results obtained based on the database using artificial intelligence algorithms. 2. The suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system as claimed in claim 1, it is characterized in that, the calculation method of described comprehensive weight comprises: evenly distribute the weights of the plurality of personal databases, and calculate the total weights of the personal databases that have different suggested results from the professional database and have the same suggested results from each other; For personal databases that are different from the professional database recommendation results but the same as each other, obtain the product of the weight of each database and the established weight proportional coefficient of the database, and use the sum of all the products as the result that is different from the professional database recommendation results and mutually recommended results The total weight of the same personal database; The average value of the two total weights obtained by implementing the above steps is taken as the comprehensive weight. 3. the suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system as claimed in claim 2, it is characterized in that, described database weight proportional coefficient is the adopted suggestion result that the corresponding database provides earlier in predetermined time period Rate. 4. The suggestion result feedback method of human-machine collaborative weak artificial intelligence cloud system as claimed in claim 3, it is characterized in that, described professional data update decision-making value is given and given by described multiple personal databases in predetermined period of time. Professional database The ratio of the number of suggested results with different and identical suggested results to the total number of suggested results. 5. The suggestion result feedback method of the human-machine collaborative weak artificial intelligence cloud system according to any one of claims 1-4, wherein the updating corresponding data of the professional database comprises: Store the question and the suggestion results given by the personal database which is different from the professional database suggestion result and the same as each other's suggestion result as professional data in the professional database or replace the original corresponding data in the professional database; Train on specialized databases with updated data.

Images