Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. This application is capable of implementation in many different ways than those herein set forth and of similar import by those skilled in the art without departing from the spirit of this application and is therefore not limited to the specific implementations disclosed below.
The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It will be understood that, although the terms first, second, etc. may be used herein in one or more embodiments to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first can also be referred to as a second and, similarly, a second can also be referred to as a first without departing from the scope of one or more embodiments of the present description. The word "if," as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination," depending on the context.
First, the noun terms relating to one or more implementations of the present invention are explained.
Mutual insurance: the unit or the individual with the homogeneous risk guarantee requirement becomes a member through contracting, and pays the premium to form a mutual fund, and the loss caused by the contract accident by the fund assumes the responsibility of compensation, or assumes the insurance activity for paying the responsibility of insurance when the insured person dies, is disabled, has diseases or reaches the conditions of the contract age, the contract deadline and the like.
In the present application, a problem processing method is provided, and the present specification relates to a problem processing apparatus, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.
FIG. 1 shows a flow diagram of a problem handling method, including steps 102 through 108, according to an embodiment of the present description.
Step 102: and extracting the key word characteristics of the question to be answered.
In one or more embodiments of the present specification, the question to be answered includes: a mutual aid project question which is provided by project members participating in the mutual aid project aiming at the mutual aid project; the answering user comprises: project members participating in the mutual aid project; the answering user set comprises: a set of item members comprised of item members participating in the mutual aid item. Wherein the mutual aid project can be a mutual aid coordination project, a sharing project or a poverty relief project, and the project members are members who successfully participate in any project; here, the problem processing method will be described by taking the mutual aid item as a mutual aid and coordination item as an example.
Based on this, the mutual aid assisting project refers to project members with the same requirements, the project members can participate in the mutual aid project to mutually assist in a mutual aid manner, and can be understood as project members participating in the mutual aid project at the same time, when any one project member of all the members is unexpected or needs help, as long as the accident or the help needing of the project member is within the accepting range of the mutual aid assisting project, other members participating in the mutual aid project can provide certain mutual aid money or certain help for the project member.
In the specific implementation, when a user participates in a mutual-aid coordination project, and the user joins the mutual-aid coordination project for the first time to become a project member, the user may not know the mutual-aid coordination project particularly, and the project member may issue a question on a platform corresponding to the mutual-aid coordination project, ask an unknown question, wait for answers of other members in the mutual-aid coordination project, and obtain an answer corresponding to the question.
When a project member puts forward a question, in order to enable the project member to quickly obtain an accurate answer, a platform corresponding to a mutual-help coordination project matches an answering project member according to the question put forward by the project member, the keyword feature of the question is extracted, the keyword feature of the question and the answering feature of the answering project member are calculated to obtain the feature similarity, the feature similarity and the answering score of the answering project member are calculated to obtain the answering matching degree, the answering project member larger than the preset matching degree threshold is determined to be the answering project member capable of answering the question, the answering project member capable of answering the question is matched according to the question, the timeliness of answering the question is improved, meanwhile, a plurality of answering project members can be distributed to answer the question, the probability of answering the question is improved, meanwhile, the answering project member high in relevance with the question is matched to answer the question, the problem with low answering quality is solved, and the project member putting forward the question can obtain a high-quality answer.
Specifically, the keyword features may be words, nouns or product names that are key in question sentences; for example, in a mutual aid program, the program members ask the question "how many are the members of the mutual aid program? "the keywords in the question can be determined as" mutual aid item "and" member ", so that it is convenient to determine what type the type of the question belongs to specifically by the keywords, and the answering process to the question can be completed more quickly.
In one or more embodiments of the present specification, in order to accurately obtain the keyword features of the question to be answered in the process of extracting the keyword of the question to be answered, the following specific implementation may be adopted:
obtaining question sentences of the questions to be answered;
performing semantic processing on the question sentences by adopting a semantic processing algorithm to obtain a keyword set after the semantic processing;
searching whether the keyword exists in a keyword library or not according to the keyword in the keyword set;
if yes, determining the keyword as the keyword feature;
if not, the keywords of the question to be answered are obtained again.
Specifically, the semantic processing algorithm is used to perform semantic processing on the question sentence, specifically, at least two keywords in the question sentence to be answered are obtained, the relevance of the at least two keywords is determined through the semantic processing algorithm, the at least two keywords are added to the keyword set when the relevance is greater than a preset relevance threshold, whether the keywords consistent with the keywords in the keyword set are stored or not is searched in the keyword library, if the keywords consistent with the keywords are used as the keyword features of the question to be answered, if the keywords do not exist, the semantic processing is performed on the question sentence to be answered again, and the keywords are returned again until the keywords consistent with the keywords existing in the keyword library are searched.
The question to be answered is taken as "how many people there are doctors in the mutual aid program? For example, the keyword features of the question to be answered are extracted and described, a semantic processing algorithm is adopted to obtain the keywords of the question, the obtained keywords comprise 'mutual aid and coordination projects', 'doctors' and 'many people', and the 'mutual aid and coordination projects' and the 'doctors' are determined to be the keyword features of the question by searching the keywords in the keyword library in the mutual aid and coordination projects.
It should be noted that the keywords in the keyword library may be screened from big data, the occurring keywords are all added to the keyword library, and the keyword library is updated in real time in order to ensure that the keywords of the question to be answered can be found in the keyword library. The storage mode of the keywords in the keyword library can be set according to practical application, and the description is not limited at all.
The key words of the questions to be answered are obtained by adopting a semantic processing algorithm, so that the essential contents of the questions to be answered can be accurately understood, and answering users with high answering accuracy can be conveniently matched in the subsequent process of matching the answering users; similarly, a word segmentation algorithm can be used to perform word segmentation on the question sentence, so as to obtain the keyword feature.
Step 104: and calculating the feature similarity between the response feature of each response user and the keyword feature according to the response feature of the response user in the response user set.
Specifically, the keyword features of the to-be-answered question are extracted according to the above, feature similarity is calculated for the keyword features and the answering features of the answering users, and the answering users suitable for answering the to-be-answered question in the answering user set can be screened out.
Based on this, the response features of the response users in the response user set may be attribute features of the response users and response keyword features of the response users, the attribute features may be occupation, gender, age and the like of the response users, and the response keyword features may be keyword features of frequently answered questions of the response users.
For example, if the answering user is a member participating in a mutual aid program and the answering user is a doctor who frequently answers questions about medical aspects, then "medical" and "doctor" may be the answering keyword features of the answering user.
In one or more embodiments of the present specification, in a case that the response feature of the response user includes an attribute feature of the response user and a response keyword feature of the response user, the feature similarity may be determined in the following manner, and a specific implementation manner is as follows:
acquiring attribute characteristics and answering keyword characteristics of each answering user in the answering user set;
according to the attribute characteristics and the answer keyword characteristics of each answer user, adopting a similarity algorithm to calculate the attribute characteristic similarity between the attribute characteristics and the keyword characteristics of each answer user, and adopting the similarity algorithm to calculate the answer characteristic similarity between the answer keyword characteristics and the keyword characteristics of each answer user;
and calculating the average value of the attribute feature similarity and the answering feature similarity to serve as the feature similarity.
Specifically, the feature similarity between the response feature and the keyword feature may be a similarity between the response feature and the keyword feature, for example, if the response feature is "traffic police" and the keyword feature is "traffic", the feature similarity between the response feature and the keyword feature determined on the basis is higher; if the answer feature is teacher and the keyword feature is tree, the similarity between the answer feature and the keyword feature determined on the basis is low.
Optionally, the similarity algorithm may be to extract semantics of the attribute features and the keyword features, determine a percentage value of similarity of the attribute features according to the semantic similarity of the attribute features and the keyword features, extract semantics of the answer keyword features and the keyword features, determine percentage data of similarity of the answer features according to the semantic similarity of the answer keyword features and the keyword features, and calculate an average value of the two percentage values to answer the feature similarity.
Taking the answering user as a project member participating in a mutual aid project as an example, describing feature similarity between the answering feature of each answering user and the keyword feature calculated according to the answering features of the answering users in an answering user set, wherein the attribute feature of the project member is ' doctor ', the answering keyword feature is ' medical ', the question to be answered is ' accident, what medical help will be provided for me in a mutual aid project? "the keyword features of the question are obtained as" provide "," medical "," mutual aid and help "according to the semantic processing algorithm, the similarity between the attribute features and the attribute features of the keyword features of the question is determined to be 60% according to the similarity algorithm, the similarity between the answer similarity features and the answer features of the keyword features of the question is determined to be 100%, and the feature similarity between the item member and the question is (60% + 100%)/2 =80%.
By determining the similarity between the answering characteristics and the keyword characteristics, answering users suitable for answering the questions to be answered can be matched for the questions to be answered corresponding to the keyword characteristics.
In one or more embodiments of the present description, in a case where the answering user is participating in a mutual aid project, the answering user set may be a project member set, and the project member set may be determined by:
acquiring historical response data of project members participating in the mutual aid project;
detecting whether the historical answering data meet preset answering conditions or not;
if yes, creating the project member set according to all project members meeting the preset answering conditions;
if not, creating a prepared project member set according to all project members which do not meet the preset answering conditions, wherein the prepared project member set has answering experiences but does not meet the preset answering conditions.
It should be noted that the preset answering condition may be a number of times of answering by the project member, a quality of answering by the project member, or a speed of answering by the project member, and the specific preset answering condition may be set according to an actual application scenario, which is not limited herein.
For example, the preset answering condition is that the number of the answering questions of the project members meets 10, the answering history of the project members participating in the mutual help project is obtained, the project member sets are created by the project members meeting the number of the answering questions greater than 10, the project members who have answered the questions but do not meet the number of the answering questions greater than 10 create a preliminary project member set, and the project members who have not answered the questions are not processed, wherein the project members in the preliminary project member set can be added to the project member set when the number of the answering questions reaches 10.
In addition, the project members in the project member set have no upper limit, and if the project members participating in the mutual aid project all meet the preset answering conditions, the project members participating in the mutual aid project all can be used as answering users.
By acquiring the attribute characteristics of the answering users and the answering keyword characteristics and calculating the characteristic similarity between the attribute characteristics and the answering keyword characteristics of the answering users and the keyword characteristics of the question to be answered, the answering users which are more suitable for answering the question to be answered can be selected for the question to be answered, and the accuracy of the answer of the question to be answered is ensured.
Step 106: and calculating the answering matching degree of each answering user and the question to be answered according to the feature similarity and the answering score of each answering user.
Specifically, according to the feature similarity obtained by the calculation, in order to select a user who is more suitable for answering the question to be answered, further calculation is performed, according to the feature similarity and the answering matching degree of each answering user, the answering matching degree of each answering user and the question to be answered is calculated, and the answering quality can be ensured by selecting the answering user with high matching degree to answer the question to be answered.
In one or more embodiments of the present specification, the answer score includes an answer quality quantum score and an answer liveness sub-score, the answer quality quantum score is used to represent past answer quality of the answer user, the answer liveness sub-score is used to represent a question that the answer user has been actively answered by a project member, and the answer score of any answer user in the answer user set can be determined in the following manner:
acquiring a historical answering record of the answering user;
calculating an answering quality sub-score and an answering activeness sub-score according to answering quality data in an answering quality dimension and answering activeness data in an answering activeness dimension contained in the historical answering records;
calculating the product of the answering quality sub-score and the corresponding quality weight coefficient and the product of the answering activity sub-score and the corresponding activity weight coefficient;
and summing the product of the response quality sub-score and the quality weight coefficient and the product of the response liveness sub-score and the liveness weight coefficient to obtain the response score.
Specifically, historical answering data of answering users are obtained, past answering records of the answering users are extracted from the historical answering data, the answering quality data and the answering activity data of the answering users are analyzed and determined according to the answering records, the answering quality data are graded to obtain answering quality quantum scores, the answering activity data are graded to obtain answering activity sub-scores, the answering quality sub-scores and the quality weight coefficients are calculated through a weighting sum algorithm, the answering activity sub-scores and the activity weight coefficients are calculated through the weighting sum algorithm, weighted sum results of the two sub-scores and the activity weight coefficients are summed, and the answering scores are determined.
In specific implementation, the response quality data may be: the answer word number of the answering user, the answering mode of the answering user, the application specific name word quantity of the answering user, the number of the answers praised by the answering user, the number of the answers commented by the answering user and the like; the response activity data may be: the answer amount of the answering user for 1 day, the answer amount of the answering user for 2 days, the answer delay of the answering user and the like.
Based on the above, the response quality sub-score and the response activity sub-score obtain the response quality sub-score by scoring each item of the response quality data, and the response activity sub-score is obtained by scoring each item of the response activity data.
For example, the response quality data is the number of response words of the response user, the scoring standard is set to be 10 points greater than 8 words, 5 points greater than 5 words, less than 8 words and 1 point less than 5 words, and the number of response words of the response user is scored according to the scoring standard to obtain the response quality quantum score.
Similarly, when the answering liveness data is the one-day answering amount of the answering user, setting the scoring standard to answer 3 questions 10 points, more than 1 question and less than 3 questions 5 points and no answer 0 point in one day, and scoring the one-day answering amount of the answering user according to the scoring standard to obtain the answering liveness sub-score.
By analogy, the response quality sub-score of the response quality data in the response quality dimension and the response activity sub-score of the response activity data in the response activity dimension can be determined in the above manner, and the description is not repeated herein.
Based on the above, according to the obtained answering quality sub-score and the answering liveness sub-score, the answering quality sub-score and the corresponding quality weight coefficient are calculated by adopting a weighting sum algorithm, the answering liveness sub-score and the corresponding liveness weight coefficient are calculated by adopting the weighting sum algorithm, and the calculation results of the two are summed to determine the answering score.
The quality weight coefficient and the liveness weight coefficient can be adjusted according to the actual application condition, the quality weight coefficient corresponding to the important answering quality sub-score can be properly adjusted to be larger, and the liveness weight coefficient corresponding to the important answering liveness sub-score can be properly adjusted to be larger.
In practical application, the answer score is still described by the above example, the number of the answer words of the answer user is 6 words, the answer quality sub-score is 5 points, the answer amount of the answer user in one day is 2 questions, the answer liveness sub-score is 5 points, wherein the quality weight coefficient corresponding to the number of the answer words of the answer user is 20%, the liveness weight coefficient corresponding to the answer amount of the answer user in one day is 10%, the result of the answer quality sub-score and the corresponding quality weight coefficient is determined to be 1.0 by a weighting sum algorithm, the result of the liveness sub-score and the corresponding liveness weight coefficient is 0.5, and the sum of the two results determines that the answer score of the answer user is 1.5.
In one or more embodiments of the present specification, the answer score of the answering user is determined by the above method, and further, the answer matching degree between the answering user and the question to be answered is calculated by the answer score and the feature similarity, and the specific implementation manner is as follows:
calculating the product of the feature similarity and the corresponding similarity weight coefficient and the product of the answering score and the corresponding answering weight coefficient;
and summing the product of the feature similarity and the similarity weight coefficient and the product of the answer score and the answer weight coefficient to obtain the answer matching degree.
Specifically, in the process of calculating the answer matching degree, the product of the feature similarity and the corresponding similarity weight coefficient and the product of the answer score and the corresponding answer weight coefficient may be calculated by using the above-mentioned weighting sum algorithm, and the results of the weighting sum algorithm of the feature similarity and the corresponding answer weight coefficient are summed to obtain the answer matching degree.
It should be noted that the similarity weight coefficient corresponding to the feature similarity and the answering weight coefficient corresponding to the answering score may be set through practical application and may be adjusted according to a specific application scenario, and this specification is not limited herein.
By adopting the weighting and algorithm to calculate the answering score of each answering user, the answering performance of each answering user in the past answering process can be more specifically known, then the answering score and the feature similarity are calculated, the answering matching degree is determined, the most suitable answering user can be further selected for the question to be answered, and the quality of the answer of the question to be answered is improved.
Step 108: and sending the answering invitation of the question to be answered to the answering users in the answering user set, wherein the answering matching degree of the answering users and the question to be answered is greater than a preset matching degree threshold value.
Specifically, the answer matching degree between each answer user and the question to be answered is obtained through calculation, the answer matching degree is compared with the matching degree threshold, the answer users larger than the matching degree threshold are determined as the answer users capable of answering the question to be answered, and answer invitations are sent to the answer users.
In specific implementation, the answering invitation can be sent to the answering user in a short message, mail or information mode in chat software, and the answering invitation carries information about the question to be answered and the best answer time, so that the answering invitation can receive the answer replied to the user who provides the question to be answered at the first time, and the experience effect of the user is improved.
In one or more embodiments of the present specification, a best batch of answering users in the answering user set are selected to answer the question to be answered, and a specific selection manner may be implemented as follows:
judging the answer matching degree of each answer user in the answer user set and the preset matching degree threshold value;
adding the answering users larger than the preset matching degree threshold value to an answering user invitation set;
and sending the answering invitations to each answering user in the answering user invitation set according to the sequence of the answering matching degrees from high to low.
Specifically, the answer matching degree of each answer user is determined through the step 106, the answer matching degree is compared with a preset matching degree threshold value, the answer users larger than the preset matching degree create an answer user invitation set, all the answer users in the answer user invitation set can answer the question to be answered, in order to enable the question to be answered to obtain the optimal answer, the answer users in the answer user invitation set can be arranged in a sequence from high to low according to the answer matching degree, and the answer user with the highest answer matching degree is preferentially sent to the answer user.
In one or more embodiments of the present specification, in order to improve the experience effect of the answering user, an answering invitation may be sent to the answering user in a time period of rest of the answering user, where before sending the answering invitation, behavior data of the answering user needs to be analyzed, and an active time period of answering of the answering user is determined according to an analysis result, which is specifically implemented as follows:
acquiring answering behavior data of the answering user participating in question answering;
and analyzing the answering behavior data, and determining the active time period of the answering user according to the analysis result.
In one or more embodiments of the present specification, sending an invitation to answer at the active time of the answering user may avoid disturbing the answering user, and a specific invitation manner is as follows:
determining answering users in the answering user set, wherein the answering matching degree of the answering users with the questions to be answered is greater than the preset matching degree threshold;
and sending the answering invitation to the answering user in the active period of the answering user.
Specifically, the answering behavior data may be the time when the answering user answered the question or the rest time of the answering user in the past, the behavior data is analyzed, the daily rest time of the user to be answered or the time of using the mobile phone may be determined according to the time when the answering user answered the question or the rest time of the answering user, an answering invitation is sent to the user to be answered at the time, and the answering user can receive the information of the answering invitation without disturbing the answering user.
In practical application, taking the answering user as a project member participating in a mutual-aid project as an example, sending an answering invitation of the to-be-answered question to an answering user with an answering matching degree with the to-be-answered question being greater than a preset matching degree threshold value in the answering user set is described, wherein the answering matching degree of the project member is 80%, the preset matching degree threshold value is 50%, the project member can be determined to meet the answering question, the project member is determined to use a mobile phone from 12 pm to 1 pm every day by collecting habits of opening mobile phone browsing of the project member every day, a platform corresponding to the mutual-aid project sends the invitation of answering the to the project member from 12 pm to 1 pm, and the to-be-answered question is displayed in information.
The question processing method provided by the specification calculates the feature similarity of the keyword features and the answering features of the answering users by extracting the keyword features of the questions to be answered, can select the answering users who are more suitable for answering the questions to be answered in the answering user set for the questions to be answered, then calculates the feature similarity and the answering scores of the answering users, can further screen the answering users selected through the feature similarity, realizes that the answering users who are more suitable for answering the questions to be answered are selected, ensures the answering quality of the questions to be answered, improves the timeliness of answering the questions to be answered, can select the active time of the answering users to send answering invitations when the answering users send the invitations to the answering users, avoids disturbing the answering users, improves the user experience effect, and can assign a plurality of answering users to answer the questions to be answered, and improves the answering probability of the questions to be answered.
The problem processing method provided in the present specification is further described below with reference to fig. 2 by taking an example of application of the problem processing method to a mutual aid project. The specific steps include steps 202 to 216.
Step 202: and acquiring the question Q proposed by the project member A.
Specifically, in the mutual aid project, the project member A who newly joins the mutual aid project is not particularly known about the mutual aid project, and the question Q is: "how many traffic policemen are in the mutual aid and help project? ".
Step 204: and extracting the keyword characteristics of the question Q.
Specifically, a semantic processing algorithm is adopted to determine how many traffic polices are in the mutual aid project? "the question Q carries out semantic processing, and the keywords of the question Q obtained after processing are: "mutual aid and coordination project" and "traffic policeman";
further, a keyword consistent with the keywords 'mutual aid and coordination project' and 'traffic police' is searched in a keyword library in the mutual aid and coordination project, the keyword 'traffic police' is determined to exist in the keyword library through searching, and then the keyword 'traffic police' is used as the keyword feature of the problem Q proposed by the project member A.
Step 206: and calculating the feature similarity of the keyword feature and the answering feature of each item member according to the answering feature of each item member in the item member set.
Specifically, the answer characteristic of each item member in the item member set takes the question characteristic with the most answer types of the item members as the answer characteristic of the item members according to the answer type of each item member contained in the historical answer record of each item member,
in the mutual aid and help project, if the question feature with the most types of 20 project members answering the questions is 'traffic police', the answer feature of the 20 project members is determined as 'traffic police', and based on the answer feature, the feature similarity between the 20 project members and the keyword feature of the question is determined to be 100% according to calculation.
Step 208: and calculating the response matching degree of each item member and the question according to the feature similarity obtained by calculation and the response score of each item member in the item member set.
For each item member in the item member set, the answer matching degree of the item member is calculated in the following mode:
answer matching = feature similarity characteristic weight coefficient + answer score as answer weight coefficient.
The method comprises the steps that the response grade of a project member is obtained through a historical response record of the project member, response quality data of the project member in a response quality dimension are obtained, response activity data in a response activity dimension are determined according to the response quality data, response activity grade is determined according to the response activity data, the response quality grade and the response activity grade are summed, and the response grade of the project member is determined;
the method for calculating the answer matching degree is adopted to calculate each project member in the 20 project members respectively, the 20 project members can be divided into three groups according to the answer matching degree of each project member obtained through calculation, the answer score of the first group of 10 project members is 10, the answer score of the second group of 5 project members is 5, the answer score of the third group of 5 project members is 0, the answer matching degree of the first group is determined to be 80% according to the calculation method of the answer matching degree, the answer matching degree of the second group is 50%, and the answer matching degree of the third group is 0%.
Step 210: judging whether the answer matching degree of each project member is greater than a preset matching degree threshold value or not; if not, go to step 212; if yes, go to step 214.
Specifically, the preset matching degree threshold is 60%.
Step 212: adding the item member to the preliminary item member invitation set.
Specifically, according to the determination in step 208 that the answer matching degrees of the second group and the item members in the third group of the 20 item members are all smaller than the preset matching degree threshold, the item members of the second group and the third group are added to the preliminary item member invitation set.
Step 214: the item member is added to the item member invitation set.
Specifically, according to the determination in step 208 that the answer matching degrees of the item members in the first group of the 20 item members are all greater than the preset matching degree threshold, the item members in the first group are added to the item member invitation set.
Step 216: and sending answer invitations to the item members in the activity periods of the item members in the item member invitation set according to the order of the answer matching degrees from high to low.
For the project members in the first group of the 20 project members, respectively determining the activity period of each project member according to the behavior data of the past one month of each project member for receiving information by using a mobile phone, and sending the answering invitation of the question Q in the activity period of each project member.
According to the question processing method provided by the specification, the similarity between the keyword characteristics and the answering characteristics of the project members is calculated by extracting the keyword characteristics of the question, the project members which are more suitable for answering the question in the project member set can be selected for the question, then the answer scores of the characteristic similarity and the project members are calculated, the project members selected through the characteristic similarity can be further screened, the project members which are more suitable for answering the question are selected, the answering quality of the question is ensured, the timeliness of answering the question is improved, the answering invitation is sent in the active time period of each project member in the project member invitation set, the project members in the project member invitation set are prevented from being disturbed, the experience effect of the project members is improved, meanwhile, a plurality of project members are assigned to answer the question, and the answering probability of the question is improved.
Corresponding to the above method embodiment, the present specification further provides an embodiment of a problem processing apparatus, and fig. 3 shows a schematic structural diagram of the problem processing apparatus according to an embodiment of the present specification. As shown in fig. 3, the apparatus includes:
an extraction module 302 configured to extract keyword features of a question to be answered;
a feature similarity calculation module 304, configured to calculate feature similarity between the response feature of each response user and the keyword feature according to the response feature of the response user in the response user set;
a matching degree calculating module 306 configured to calculate a matching degree of each answering user with the question to be answered according to the feature similarity and the answering score of each answering user;
the sending module 308 is configured to send an answering invitation of the question to be answered to the answering users in the answering user set whose answering matching degree with the question to be answered is greater than a preset matching degree threshold.
In an optional embodiment, the sending module 308 includes:
the judging unit is configured to judge the answer matching degree of each answer user in the answer user set and the preset matching degree threshold value;
an adding unit configured to add the answering user who is greater than the preset matching degree threshold to an answering user invitation set;
and the sending unit is configured to send the answering invitation to each answering user in the answering user invitation set according to the sequence of the answering matching degree from high to low.
In an optional embodiment, the answer score of each answering user is determined by:
aiming at any one answering user in the answering user set, the following operations are executed:
the obtaining module is configured to obtain a historical answering record of the answering user;
the first calculation module is configured to calculate sub-scores of the answering quality and sub-scores of the answering liveness according to the answering quality data in the answering quality dimension and the answering liveness data in the answering liveness dimension contained in the historical answering records;
a second calculation module configured to calculate a product of the answer quality sub-score and a corresponding quality weight coefficient, and a product of the answer liveness sub-score and a corresponding liveness weight coefficient;
a summation module configured to sum the product of the response quality sub-score and the quality weight coefficient and the product of the response liveness sub-score and the liveness weight coefficient as the response score.
In an optional embodiment, the module 306 for calculating the matching degree of answers includes:
a calculation unit configured to calculate a product of the feature similarity and a corresponding similarity weight coefficient, and a product of the response score and a corresponding response weight coefficient;
a summing unit configured to sum the product of the feature similarity and the similarity weight coefficient and the product of the answer score and the answer weight coefficient as the answer matching degree.
In an optional embodiment, before the sending module 308 operates, the method further includes:
the data acquisition response behavior module is configured to acquire response behavior data of the response user participating in question response;
and the determining module is configured to analyze the answering behavior data and determine the activity period of the answering user according to the analysis result.
In an optional embodiment, the sending module 308 includes:
a determining unit configured to determine answering users in the answering user set, wherein the answering matching degree of the answering users with the questions to be answered is greater than the preset matching degree threshold;
a sending answering invitation unit configured to send the answering invitation to the answering user during the answering user's active period.
In an alternative embodiment, the module 304 for calculating feature similarity includes:
the answer keyword feature acquisition unit is configured to acquire attribute features and answer keyword features of each answer user in the answer user set;
a first calculation similarity unit configured to calculate, according to the attribute feature and the answer keyword feature of each answer user, an attribute feature similarity between the attribute feature and the keyword feature of each answer user by using a similarity algorithm, and calculate an answer feature similarity between the answer keyword feature and the keyword feature of each answer user by using the similarity algorithm;
a second calculation similarity unit configured to calculate an average value of the attribute feature similarity and the answer feature similarity as the feature similarity.
In an optional embodiment, the extracting module 302 includes:
an acquisition question sentence unit configured to acquire a question sentence of the question to be answered;
the semantic processing unit is configured to perform semantic processing on the question sentences by adopting a semantic processing algorithm to obtain a keyword set after the semantic processing;
the searching unit is configured to search whether the keyword exists in a keyword library according to the keyword in the keyword set;
if yes, operating and determining a keyword feature unit;
the keyword feature unit is configured to determine the keyword as the keyword feature.
In an optional embodiment, the question to be answered includes: a mutual aid project question which is provided by project members participating in the mutual aid project aiming at the mutual aid project;
the answering user comprises: project members participating in the mutual aid project;
the answering user set comprises: a set of item members comprised of item members participating in the mutual aid item.
In an alternative embodiment, the set of item members is determined by:
the acquisition history response data module is configured to acquire history response data of project members participating in the mutual aid project;
the detection module is configured to detect whether the historical answering data meet preset answering conditions;
if yes, operating the creating module;
the creating module is configured to create the project member set according to all project members meeting the preset answering conditions.
The problem processing device provided by this specification, through extracting the keyword characteristic of the question of waiting to answer, calculate the characteristic similarity of keyword characteristic with the user's of answering characteristic, can be for the user selects to answer relatively is fit for answering in the user set of waiting to answer the question the user of answering of waiting to answer, rethread calculation the characteristic similarity with the user's of answering score, can go on further selection to the user who answers through the selection of characteristic similarity, realized selecting and be more fit for answering the user who answers the question of waiting to answer, guaranteed to the answer quality of the question of waiting to answer, improved the timeliness of answering the question of waiting to answer, when sending the user of answering to the answer, can select to be in the active time of the user of answering sends the invitation to answer, avoid disturbing the user of answering, improved user experience effect, can assign a plurality of users to answer the question of waiting to answer simultaneously, improved the probability that the question of waiting to answer is answered.
The above is a schematic scheme of a problem handling apparatus of the present embodiment. It should be noted that the technical solution of the problem processing apparatus and the technical solution of the problem processing method described above belong to the same concept, and details that are not described in detail in the technical solution of the problem processing apparatus can be referred to the description of the technical solution of the problem processing method described above.
Fig. 4 shows a block diagram of an electronic device 400 according to an embodiment of the present description. The components of the electronic device 400 include, but are not limited to, a memory 410 and a processor 420. Processor 420 is coupled to memory 410 via bus 430 and database 450 is used to store data.
Electronic device 400 also includes access device 440, access device 440 enabling electronic device 400 to communicate via one or more networks 460. Examples of such networks include the Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the internet. Access device 440 may include one or more of any type of network interface (e.g., a Network Interface Card (NIC)) whether wired or wireless, such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, a worldwide interoperability for microwave access (Wi-MAX) interface, an ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a bluetooth interface, a Near Field Communication (NFC) interface, and so forth.
In one embodiment of the present description, the other components of the electronic device 400 described above and not shown in fig. 4 may also be connected to each other, for example, by a bus. It should be understood that the block diagram of the electronic device shown in fig. 4 is for exemplary purposes only and is not intended to limit the scope of the present disclosure. Those skilled in the art may add or replace other components as desired.
The electronic device 400 may be any type of stationary or mobile electronic device, including a mobile computer or mobile electronic device (e.g., tablet, personal digital assistant, laptop, notebook, netbook, etc.), a mobile phone (e.g., smartphone), a wearable electronic device (e.g., smartwatch, smartglasses, etc.), or other type of mobile device, or a stationary electronic device such as a desktop computer or PC. The electronic device 400 may also be a mobile or stationary server.
Wherein processor 420 is configured to execute the following computer-executable instructions:
according to a first aspect of embodiments herein, there is provided a problem handling method, including:
extracting the key word characteristics of the question to be answered;
calculating the feature similarity between the response features of each response user and the keyword features according to the response features of the response users in the response user set;
calculating the answering matching degree of each answering user and the question to be answered according to the feature similarity and the answering score of each answering user;
and sending the answering invitation of the question to be answered to the answering users in the answering user set, wherein the answering matching degree of the answering users and the question to be answered is greater than a preset matching degree threshold value.
Optionally, the sending an invitation to answer the question to the answering user in the answering user set whose answer matching degree with the question to be answered is greater than a preset matching degree threshold includes:
judging the answer matching degree of each answer user in the answer user set and the size of the preset matching degree threshold;
adding the answering users larger than the preset matching degree threshold value to an answering user invitation set;
and sending the answering invitation to each answering user in the answering user invitation set according to the sequence of the answering matching degree from high to low.
Optionally, the answer score of each answering user is determined by:
aiming at any one answering user in the answering user set, the following operations are executed:
acquiring a historical answering record of the answering user;
calculating an answering quality sub-score and an answering activeness sub-score according to answering quality data in an answering quality dimension and answering activeness data in an answering activeness dimension contained in the historical answering records;
calculating the product of the answering quality sub-score and the corresponding quality weight coefficient and the product of the answering activity sub-score and the corresponding activity weight coefficient;
and summing the product of the response quality sub-score and the quality weight coefficient and the product of the response liveness sub-score and the liveness weight coefficient to obtain the response score.
Optionally, the calculating, according to the feature similarity and the answer score of each answering user, an answer matching degree between each answering user and the question to be answered includes:
calculating the product of the feature similarity and the corresponding similarity weight coefficient and the product of the answer score and the corresponding answer weight coefficient;
and summing the product of the feature similarity and the similarity weight coefficient and the product of the answer score and the answer weight coefficient to obtain the answer matching degree.
Optionally, before the step of sending an invitation to answer the question to the answering user in the answering user set whose answer matching degree with the question to be answered is greater than a preset matching degree threshold is executed, the method further includes:
acquiring answering behavior data of the answering user participating in question answering;
and analyzing the answering behavior data, and determining the active time period of the answering user according to the analysis result.
Optionally, the sending an invitation to answer the question to the answering user in the answering user set whose answer matching degree with the question to be answered is greater than a preset matching degree threshold includes:
determining answering users in the answering user set, wherein the answering matching degree of the answering users and the questions to be answered is greater than the preset matching degree threshold;
and sending the answering invitation to the answering user in the active period of the answering user.
Optionally, the calculating, according to the response characteristics of the response users in the response user set, the feature similarity between the response characteristics of each response user and the keyword features includes:
acquiring attribute characteristics and answering keyword characteristics of each answering user in the answering user set;
according to the attribute characteristics and the answer keyword characteristics of each answer user, adopting a similarity algorithm to calculate the attribute characteristic similarity between the attribute characteristics and the keyword characteristics of each answer user, and adopting the similarity algorithm to calculate the answer characteristic similarity between the answer keyword characteristics and the keyword characteristics of each answer user;
and calculating the average value of the attribute feature similarity and the answering feature similarity as the feature similarity.
Optionally, the extracting the keyword feature of the question to be answered includes:
obtaining question sentences of the questions to be answered;
performing semantic processing on the question sentences by adopting a semantic processing algorithm to obtain a keyword set after the semantic processing;
searching whether the keyword exists in a keyword library or not according to the keyword in the keyword set;
and if so, determining the keyword as the keyword feature.
Optionally, the question to be answered includes: a mutual aid project problem which is provided by project members participating in a mutual aid project aiming at the mutual aid project;
the answering user comprises: project members participating in the mutual aid project;
the answering user set comprises: a set of item members comprised of item members participating in the mutual aid item.
Optionally, the set of item members is determined by:
acquiring historical response data of project members participating in the mutual aid project;
detecting whether the historical answering data meet preset answering conditions or not;
and if so, creating the project member set according to all project members meeting the preset answering conditions.
The foregoing is a schematic scheme of an electronic device of this embodiment. It should be noted that the technical solution of the electronic device and the technical solution of the problem handling method belong to the same concept, and details that are not described in detail in the technical solution of the electronic device can be referred to the description of the technical solution of the problem handling method.
An embodiment of the present application also provides a computer-readable storage medium storing computer instructions, which when executed by a processor, implement the steps of the problem handling method as described above.
The above is an illustrative scheme of a computer-readable storage medium of the present embodiment. It should be noted that the technical solution of the storage medium belongs to the same concept as the technical solution of the problem handling method, and details that are not described in detail in the technical solution of the storage medium can be referred to the description of the technical solution of the problem handling method.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The computer instructions comprise computer program code which may be in the form of source code, object code, an executable file or some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM), random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer-readable medium may contain suitable additions or subtractions depending on the requirements of legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer-readable media may not include electrical carrier signals or telecommunication signals in accordance with legislation and patent practice.
It should be noted that, for the sake of simplicity, the above-mentioned method embodiments are described as a series of acts or combinations, but those skilled in the art should understand that the present application is not limited by the described order of acts, as some steps may be performed in other orders or simultaneously according to the present application. Further, those skilled in the art will appreciate that the embodiments described in this specification are presently considered to be preferred embodiments and that acts and modules are not required in the present application.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
The preferred embodiments of the present application disclosed above are intended only to aid in the explanation of the application. Alternative embodiments are not exhaustive and do not limit the invention to the precise embodiments described. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the application and the practical application, to thereby enable others skilled in the art to best understand and utilize the application. The application is limited only by the claims and their full scope and equivalents.