CN109344227A - Worksheet method, system and electronic equipment - Google Patents
Worksheet method, system and electronic equipment Download PDFInfo
- Publication number
- CN109344227A CN109344227A CN201810682171.5A CN201810682171A CN109344227A CN 109344227 A CN109344227 A CN 109344227A CN 201810682171 A CN201810682171 A CN 201810682171A CN 109344227 A CN109344227 A CN 109344227A
- Authority
- CN
- China
- Prior art keywords
- work order
- type
- answer
- classification
- order problem
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of worksheet methods, system and electronic equipment, it is related to work order to report an error the technical field of processing, including the counsel requests by receiving reporting of user, work order problem is carried in counsel requests, one of the following contents: menu code is included at least in work order problem, transaction code, report an error code and the description of work order problem, the problem of determining work order problem type, problem types include first problem type and Second Problem type, first problem type be include one of the following contents menu code, transaction code, the problem of reporting an error yard type, Second Problem type is to include at least the problem of work order problem describes type, based on problem types, it obtains recommending answer, it is pushed to user, the present invention is by classifying work order problem, and it is based on different problems type, obtain push answer, it is pushed to user, improve the matching essence of answer Degree, enhances the experience sense of user.
Description
Technical field
It reports an error processing technology field the present invention relates to work order, more particularly, to a kind of worksheet method, system and electronics
Equipment.
Background technique
Many menus are contained in one operation system, each menu can issue multiple service requests, when staff makes
When with the operation system, when mistake occurs for some service request of some menu, it is if the business personnel thinks the mistake
Normally report an error or can not understand when reporting an error, this problem that reports an error will be submitted to worksheet system queries answer.
Existing worksheet system generally can be according to the keyword query answer for the problem of reporting an error.However, by business personnel
The problem of reporting an error reported is diversified, such as: it may be that menu code, and/or person's transaction code, and/or person report an error the report of code
Wrong problem, the problem that reports an error that can also be described only for a problem, so, simply by virtue of the keyword query for the problem of reporting an error
Answer, often very low for the matching degree of answer, user experience is not high.
Summary of the invention
In view of this, improving answer the purpose of the present invention is to provide worksheet method, system and electronic equipment
Matching precision enhances the experience sense of user.
In a first aspect, the embodiment of the invention provides a kind of worksheet methods, comprising:
The counsel requests for receiving reporting of user, carry work order problem in the counsel requests, in the work order problem extremely
Less include one of the following contents: menu code, transaction code, report an error code and the description of work order problem;
The problem of determining work order problem type;Described problem type includes first problem type and Second Problem class
Type;The first problem type is type the problem of including at least one of the following contents menu code, transaction code, report an error yard, described
Second Problem type is to include at least the problem of work order problem describes type;
Based on described problem type, obtains recommending answer, be pushed to the user.
With reference to first aspect, the embodiment of the invention provides the first possible embodiments of first aspect, wherein base
In described problem type, obtains recommending answer, is pushed to the user, comprising:
Determine described problem type be first problem type after, compared from trigram answer library whether have solution with
The first problem type matches;
If it is, extract the solution to match with the first problem type, using the solution as
Answer is recommended to be pushed to the user.
With reference to first aspect, the embodiment of the invention provides second of possible embodiments of first aspect, wherein base
In described problem type, obtains recommending answer, is pushed to the user, comprising:
After determining that described problem type is Second Problem type, content characteristic is extracted;The content characteristic is from work order
It is extracted in work order problem description in problem;
Based on the content characteristic, classification belonging to the work order problem is determined;Each classification is corresponding with classification answer
Library;
By the work order problem one by one each of classification answer library corresponding with classification storage work order problem into
Row similarity-rough set;Classification answer library includes at least a storage work order problem;
Determine that the corresponding storage work order solution to the problem of similarity maximum value to recommend answer, is pushed to the use
Family.
With reference to first aspect, the embodiment of the invention provides the third possible embodiments of first aspect, wherein mentions
Take content characteristic, comprising:
Multiple text participles are extracted from the work order problem description in work order problem;
Using TF-IDF, the content characteristic is extracted from the multiple text participle.
With reference to first aspect, the embodiment of the invention provides the 4th kind of possible embodiments of first aspect, wherein
After extracting the step of multiple texts segment in the work order problem description from work order problem, TF-IDF is utilized described,
Before the step of extracting the content characteristic in the multiple text participle, the method also includes:
To the multiple original text participle be filtered processing, with filter out low-frequency word, stop words, mark information, together
Adopted word.
With reference to first aspect, the embodiment of the invention provides the 5th kind of possible embodiments of first aspect, wherein base
In the content characteristic, classification belonging to the work order problem is determined, comprising:
Determine the work order type of service of the work order problem;
According to the work order type of service, reverse classifier is obtained;
The content characteristic is input in the reverse classifier, determines classification belonging to the work order problem.
With reference to first aspect, the embodiment of the invention provides the 6th kind of possible embodiments of first aspect, wherein institute
Reverse classifier is stated to obtain by the following method:
The content characteristic of classification and multiple storage work order problems to multiple storage work order problems carries out automatic cluster, obtains
Reverse classifier;
And/or manual command is received, it is based on the manual command, optimizes the reverse classifier.
With reference to first aspect, the embodiment of the invention provides the 7th kind of possible embodiments of first aspect, wherein will
Each of classification answer library corresponding with classification storage work order problem carries out similarity ratio to the work order problem one by one
Compared with, comprising:
Obtain the storage content characteristic of each of classification answer library storage work order problem;
Based on the cosine law, the content characteristic and the storage content characteristic are subjected to similarity-rough set.
Second aspect, the embodiment of the present invention also provide a kind of worksheet system, comprising:
Receiving unit carries work order problem for receiving the counsel requests of reporting of user in the counsel requests, described
One of the following contents is included at least in work order problem: menu code, transaction code, report an error code and the description of work order problem;
Determination unit, type the problem of for determining the work order problem;Described problem type includes first problem type
With Second Problem type;The first problem type is class the problem of including one of the following contents menu code, transaction code, report an error yard
Type, the Second Problem type are to include at least the problem of work order problem describes type;
Push unit obtains recommending answer, is pushed to the user for being based on described problem type.
The third aspect, the embodiment of the present invention also provide a kind of electronic equipment, including memory, processor, the memory
In be stored with the computer program that can be run on the processor, the processor is realized when executing the computer program
The step of stating embodiment described in any item methods.
The embodiment of the present invention brings following the utility model has the advantages that passing through the counsel requests for receiving reporting of user, in counsel requests
Work order problem is carried, include at least one of the following contents in work order problem: menu code, transaction code, report an error code and work order problem
The problem of describing, determining work order problem type, problem types include first problem type and Second Problem type, first problem class
Type is type the problem of including one of the following contents menu code, transaction code, report an error yard, and Second Problem type is to include at least work order
The problem of problem describes type is based on problem types, obtains recommending answer, is pushed to user, and the present invention is by by work order problem
Classify, and be based on different problems type, obtains push answer, be pushed to user, improve the matching precision of answer, increase
The strong experience sense of user.
Other features and advantages of the present invention will illustrate in the following description, also, partly become from specification
It obtains it is clear that understand through the implementation of the invention.The objectives and other advantages of the invention are in specification, claims
And specifically noted structure is achieved and obtained in attached drawing.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art be briefly described, it should be apparent that, it is described below
Attached drawing is some embodiments of the present invention, for those of ordinary skill in the art, before not making the creative labor
It puts, is also possible to obtain other drawings based on these drawings.
Fig. 1 is the flow chart of worksheet method provided by one embodiment of the present invention;
Fig. 2 is the flow chart for the worksheet method that another embodiment of the present invention provides;
Fig. 3 is the flow chart of the worksheet method provided in an embodiment of the present invention in Second Problem type;
Fig. 4 is the flow chart of machine learning provided in an embodiment of the present invention;
Fig. 5 is work order problem to be measured provided in an embodiment of the present invention and the schematic diagram that work order Question Classification compares;
Fig. 6 is the flow chart of machine autonomous learning provided in an embodiment of the present invention;
Fig. 7 is the schematic diagram of reverse classifier training provided in an embodiment of the present invention;
Fig. 8 is the schematic diagram of reverse classifier building and work order problem identification provided in an embodiment of the present invention;
Fig. 9 is the schematic diagram that work order provided in an embodiment of the present invention is categorized into multitiered network;
Figure 10 is the schematic diagram provided in an embodiment of the present invention identified using multistratum classification network to work order problem;
Figure 11 be it is provided in an embodiment of the present invention based on type of service to the flow chart of the maintenance of content characteristic;
Figure 12 is the flow chart of content characteristic provided in an embodiment of the present invention and the maintenance of answer mapping relations;
Figure 13 is the structure chart of worksheet system provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Technical solution be clearly and completely described, it is clear that described embodiments are some of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
For convenient for understanding the present embodiment, first to a kind of worksheet method disclosed in the embodiment of the present invention into
Row is discussed in detail,
As shown in connection with fig. 1, a kind of flow chart of worksheet method is shown, which comprises
S110: receiving the counsel requests of reporting of user, carries work order problem in the counsel requests, in work order problem extremely
Less include one of the following contents: menu code, transaction code, report an error code and the description of work order problem;
It specifically,, should when mistake occurs for some service request of some menu when user uses the operation system
User think the mistake be normally report an error or can not understand when reporting an error, this problem that reports an error (counsel requests) will be submitted to work
In uniprocesser system, at this moment, worksheet system will receive the counsel requests of reporting of user.In this case, consulting is asked
Carry work order problem in asking, may include in work order problem: menu code, and/or person's transaction code, and/or person report an error code.
In addition, this query can be submitted to worksheet system when user has a question to some business to seek advice from, used
Family can also report counsel requests to worksheet system, at this moment, in counsel requests carry work order problem, which can
To only include the description of work order problem.
S120: the problem of determining work order problem type;Problem types include first problem type and Second Problem type;The
One problem types are types the problem of including at least one of the following contents menu code, transaction code, report an error yard, and Second Problem type is
The problem of being described including at least work order problem type;
Specifically, it when including at least one of the following contents menu code, transaction code in work order problem, reporting an error yard, determines
The problem of work order problem type is first problem type, and when describing in work order problem including at least work order problem, determining should
The problem of work order work order problem type is Second Problem type.
Specific method of determination can be that can detecte whether work order problem encodes first, if there is coding, then can determine
It reports an error code in work order problem comprising menu code, and/or person's transaction code, and/or person.It, can be with if work order problem does not encode
The problem of determining work order problem type is Second Problem type.
S130: being based on problem types, obtains recommending answer, is pushed to user.
As shown in connection with fig. 2, when user submits work order problem, it is first determined whether work order problem includes trigram, wherein three
Code includes menu code, transaction code, report an error code.If so, can then go matching answer from trigram answer library, output is recommended to answer
Case.If it is not, then needing to carry out semantic analysis, go to match answer in category database, answer is recommended in output.
Specifically for Fig. 2, determine problem types be first problem type after, compared from trigram answer library whether
There is solution to match with first problem type;
If it is, the solution to match with first problem type is extracted, using solution as recommendation answer
It is pushed to user.
Specifically, when user's submission system report an error problem when, carry in the problem that reports an error and currently report an error the three of interface
Code information, and be compared with trigram answer library.When trigram answer library compares successfully, problem has solution, and system is straight
It connects and returns to solution immediately in current interface, is i.e. user is given in recommendation answer.When trigram answer library compares failure, i.e. system
Do not provide solution to the problem, user can supplement according to the content of the answer of last time improve work order problem after submit consult
Request is ask, user can lift the secondary counsel requests of work order in current interface.
In addition, the identification of the trigram based on work order, can carry out the batch processing of similar work order.
It by work order problem and does not currently solve mark post work order problem and carries out trigram comparison;It compares successfully, and same problems are
There are other employees to propose, automatically sort out two work order problems, after mark post work order solution to the problem determines, presses
Same scheme batch solves.Wherein, mark post work order problem is to confirm that the solution to the problem is standard in trigram answer library
The work order problem of solution.
Fail when comparing, and same problems there is no solution, sets mark post work order for the work order problem automatically and asks
Topic, to sort out the subsequent submission work order of processing.
Special hilllock personnel can manually adjust the mark post work order problem being arranged automatically;Have under such class mark post work order problem
What work order was sorted out, a key replacement can be carried out to mark post work order problem.
Trigram answer library can be managed during establishing with final-period management by way of guided learning and update.
Specifically, after the mark post work order problem with " trigram " information obtains solution, work order is asked in intelligent Kuku
Topic is oriented study, extracts " trigram " information and solution, after special hilllock personnel adjustment, confirmation, is included in trigram answer library.
" trigram " information and the existing content in trigram answer library extracted during guided learning are duplicate, and system carries out automatically
Prompt;After special hilllock staff integration, confirmation, trigram answer library is updated.
Specifically for Fig. 2, after determining that problem types are Second Problem type, in conjunction with shown in Fig. 3,
S310: content characteristic is extracted;Content characteristic is to extract from the work order problem description in work order problem;
Step S310 includes:
Multiple text participles are extracted from the work order problem description in work order problem;
To the multiple original text participle be filtered processing, with filter out low-frequency word, stop words, mark information, together
Adopted word;
Using TF-IDF, content characteristic is extracted from multiple texts participle Jing Guo filtration treatment.
Specifically, work order problem is handled, first has to indicate to become by content of text mathematically analyze place
The form of reason, that is, establish text feature, characterizes text information with certain characteristic item.That is, extracting in work order problem first
Multiple text participles, and have part-of-speech tagging and unknown word identification.
After the word segmentation is completed, it is also necessary to text participle collection be performed some processing, useless information is filtered out, to reduce text
The dimension of the vector of eigen composition, reduces unnecessary operation.Specifically filtration treatment includes:
(1) filtering low word, such as certain words only occurred in one or two of text, these words are stayed in set and can be led
The attribute value for causing most of text is 0.In view of text classification, the word that word frequency is 1 is only filtered out.
(2) stop words is filtered, general this word hardly carries any information.Such as: " ", " ground ", " obtaining " these help
Word, and " since ", the word of the reflection such as " still " syntactic structure, they can neither reflect the semanteme of text, can also be to key
The extraction of word interferes, and needs to be filtered.It establishes and deactivates dictionary, include all function words and punctuation mark.
(3) passport NO., account, cell-phone number data information in mark information, such as text are filtered.
(4) synonym is filtered, thesaurus is established, it is synonym that such as " payment ", which is " payment ", if wrapped in text
Containing " payment ", then synonymous standard words " payment " are converted into.
After completing multiple text participle filterings, using TF-IDF, content characteristic is extracted from multiple texts participle.
Specifically, text feature is sought to calculate the result of participle, common calculation method has TF-
IDF.Wherein TF is known as word frequency, the ability for describing document content for characterizing the word;IDF is known as anti-document frequency, for characterizing
The ability of word differentiation document.Text feature includes:
(1) word frequency is calculated
Text is divided into length, and for the ease of the comparison of different texts, word frequency needs to standardize
It wherein, is the frequency of occurrence of word in a document, wdFor total word number in document.
(2) inverse document frequency (the characterization ability of word) is calculated
IDF (t)=log (N/nc+1)
It is the textual data that word occurs in the training set of answer library for the textual data of answer library training set.
(3) TF-IDF is calculated
TFIDF (t, d)=TF (t, d) * IDF (t, d)
TF-IDF is directly proportional to the frequency of occurrence of a word in a document, with frequency of occurrence of the word in entire language at
Inverse ratio.The weighted value that Feature Words are calculated with TF-IDF algorithm is to indicate that the frequency occurred in this document when a word is got over
Height, while the number occurred in other documents is fewer, then shows that the word is stronger for the separating capacity for indicating the document, so
Its weighted value just should be bigger.The weight of all words is sorted, it as needed can be there are two types of selection mode:
(1) n keyword of the maximum a certain fixed number of weight is selected;
(2) selection weight is greater than the keyword of a certain threshold value.
First way is used in the present invention, and sets 1000 for n value, and the text of each answer or problem is corresponding
The space vector of one 1000 dimension, i.e. feature vector.Each feature vector identify a text, behind will be right by these
Feature vector calculates similarity and text cluster.
S320: being based on content characteristic, determines classification belonging to the consulting of work order problem;Each classification is corresponding with classification answer
Library;
Specifically,
Determine the work order type of service in work order problem;
According to work order type of service, reverse classifier is obtained;
Content characteristic is input in reverse classifier, determines classification belonging to work order problem.
Wherein, reverse classifier obtains by the following method:
The content characteristic of classification and multiple storage work order problems to multiple storage work order problems carries out automatic cluster, obtains
Reverse classifier;
And/or manual command is received, it is based on manual command, optimizes reverse classifier.
Specifically, in the present invention, as shown in connection with fig. 4, machine can be carried out to the work order library of existing storage work order problem
Then device study, building category feature library, answer library and its mapping relations are carried out by the work order problem to be measured that work order consulting reports
It characterizes and extracts, compared with category feature library, answer of patrolling in answer library.
The main means of step S320 are in the work order problem (work order problem hereinafter referred to as to be measured) of reporting of user and library
Settled storage work order problem is compared, and finds immediate storage work order problem in work order problem to be measured and library, will most
The close corresponding work order answer of storage work order problem recommends user as recommendation answer.
However, since work order data volume is big, and increase with time, so if by all works in work order problem to be measured and library
Single problem is compared, then needs extremely huge calculating, does not have application value.In order to reduce number of comparisons, matching is improved
Efficiency needs to classify to storage work order problem as shown in connection with fig. 5, by work order problem it is similar be classified as one kind.In this way, first
First judge that classification belonging to work order problem to be measured, later period, work order problem to be measured only need storage work order corresponding with each classification
Problem is compared can find classification described in work order problem to be measured, multiple tools in classification in storage work order problem
Most similar storage work order problem is found in the storage work order problem for having answer, it will most similar storage work order problem be corresponding answers
Case is pushed to user as recommendation answer.
Since classification number is much smaller than chief engineer's odd number, so first classify when finding most like storage work order problem,
So that the number that the later period compares effectively reduces, system identification efficiency is greatly improved.Due to by classifying by similar work order collection
In in one, avoid the ambiguity in identification process.
In addition, classification and category feature extraction are carried out according to work order problem in the present invention, and the quantity of work order is to be continuously increased
, so needing to carry out self-teaching, work order problem to be measured is covered in classification system.As shown in connection with fig. 6, by comparison to
It surveys work order problem and the feature of classification work order problem compares, if similarity is higher than threshold value, work order problem to be measured is divided
It is updated to known classification, and to the classification that it belongs to, if cannot be harmonious with known classification, one can be created
Classification, so the cost of autonomous learning is the growth of work order problem category number.
For how to classify to work order problem, the present invention classifies to work order problem using reverse classifier.Its
In, the Training strategy of reverse classifier are as follows: as shown in connection with fig. 7, during classifier training, first determine answer (certain classification),
Category feature is determined again, wherein answer can correspond to multiple work order problems, for example, the answer of A class can be asked with corresponding A 1~An class work order
Topic, i.e., when user submits A1~An class work order problem, the recommendation answer for recommending user is A class answer.So first answering
Case carries out feature extraction and classifying, determines the model answer of answer classification and each classification, then corresponding to same class answer again
Work order problem carry out feature extraction, determine category feature.By this method train come classifier, it is special with problem is directly passed through
It levies the classifier classified to compare, granularity is thinner, and precision is higher.
Wherein, the essence of classifier training is a unsupervised cluster process, and detailed process is as follows:
1, assume the storage work order problem log for currently thering is [G1, G2, G3..., Gm] total n item to have answer, then represent current
There are m classification, corresponding set [G1, G2 ..., Gm] in library;
2, optional work order problem Gx, to its answer carry out feature extraction obtain answer Expressive Features set [DTx1,
DTx2 ..., DTxo], which represents the current answer characteristic set of such Cx;
3, now define similarity function fn (Cx, Gy)=| | Gy-Cx | | and similarity threshold miu, if a work order is asked
The similarity for inscribing Gy and class Cx is greater than threshold value miu, then determines that work order problem Gy belongs to Cx class, otherwise Gy still falls within separate class;
If 4, work order problem Gy belongs to Cx class, using work order problem Gy answer characteristic set [DTy1,
DTy2 ..., DTyp] the original Cx class of amendment answer characteristic set [DTx1, DTx2 ..., DTxo], obtain answering for new Cx class
Pattern characteristics set [DTx1, DTx2 ..., DTxo, DTxo+1 ... DTyq];
5, by judging whether all work order problems [G1, G2, G3..., Gm] belong to class C1, new category set is obtained
[C1,G1,....,Gm-n1];
6, to remaining work order problem [G1 ..., Gm-n], repeat the above steps 1-5, obtain new category set [C1,
C2,G1,....,Gm-n1-n2];
7, step 6 is repeated, new category set [C1, C2 ..., Ck] (k≤m) is obtained.
Pattern characteristics are answered by above-mentioned reverse classifier training strategy to classify, and extract answer category feature and corresponding
After alternative answer, the classification results based on answer feature are needed, feature extraction is carried out to the corresponding work order problem of same class answer
And classification, obtain the higher work order content characteristic class of cohesion.The cohesion of feature class is higher, as shown in connection with fig. 8, shows inverse
Relationship between matching to classifier with work order problem, will be more accurate to the identification of new work order problem, and the answer of recommendation is got over
Accurately.
Similarly, during such answer corresponding work order Content Feature Extraction, identical Text character extraction is also used
Method and similarity calculating method can significantly improve work order accuracy of identification, improve the accuracy for recommending answer.
However, for a work order problem, the content in work order problem includes many elements, in work order question attributes
Include level-one type of service and secondary traffic type, can be directly used for the foundation of work order Question Classification.So classifying in design
When device, by the way of hierarchical classification, this mainly includes two aspects:
1, the classification of work order as shown in connection with fig. 9, is shown, when using storage work order problem training classifier, to storage
Work order is first classified according to a secondary traffic type, is then directed to each group, is recycled above-mentioned reverse classifier training
Strategy is trained, and obtains more accurate classification.Pass through this method, so that it may obtain the classification net of a 3 etale topology structures
Network.
2, it as shown in connection with fig. 10, illustrates how to determine classification belonging to work order problem, when needing to push away for new work order problem
When recommending answer, judgement of the work order problem Jing Guo feature extraction and 3 above-mentioned layer networks, finally belonging to the available work order problem
Classification, quickly recommend answer results best out.
The work order problematic amount of WorkForm System gradually increases over time, corresponding answer automatic searching system
Also must be one can with autonomous learning, can make accurate judgment to new work order problem and answer recommend system.So working as work
After thering is a new work order problem to be archived in single system, the content characteristic and answer feature according to the work order problem are needed, more
New existing answer library information:
If 1, the answer of the work order problem is consistent with the answer of existing work order, the corresponding content characteristic of the answer is updated
Set;
If 2, the answer of the work order problem and the answer of existing work order problem are inconsistent, increase newly answer classification with
And corresponding content characteristic set.
However, through the above steps, the automatic cluster to work order answer may be implemented.According to the automatic cluster of answer feature
Carry out structural classification device and still have following problems:
(1) each work order problem has corresponding answer, even if the answer physical meaning of two work order problems is consistent,
But due to the text of description difference, it will lead to and obtain entirely different Feature Words when carrying out feature extraction to answer, machine is being learned
Habit is possible in the process will be using the two work order problems as the work order of two classifications.
(2) being assigned to a kind of answer has many, machine can not intelligent comprehensive optimization such issues that provide optimal answering
Case.
For both of these problems, in machine-learning process, needs to be added some manual interventions, receive manual command, base
In manual command, optimize reverse classifier, so as to improve classification and problem answers accuracy.
Manual intervention mechanism is the supplement to machine learning and Intelligent treatment mechanism, and intelligent treatment process can be briefly described
According to the content of work order problem, to be matched to its corresponding content characteristic class, closed further according to the mapping of content characteristic class and answer
System, finds this corresponding answer of work order problem.For the important link during this, manual intervention mechanism is devised, is passed through
The result of machine learning is finely adjusted, to guarantee to reach optimal matching effect.Manual intervention mainly includes following
Aspect: content characteristic maintenance, answer maintenance and content characteristic and answer mapping relations are safeguarded.
Content characteristic is safeguarded: the positioning for the content characteristic class of work order problem, by the way of hierarchical classification.For
Guarantee the rapidly and efficiently retrieval for content characteristic set, it will be right on the basis of the I and II type of service of work order problem
Classify in feature, after a certain certain types of work order is submitted, is retrieved just for the character subset of this type
Matching.Content characteristic belonging to it can be navigated to using type of service as KEY in the time complexity of (1) O by being equivalent to
Collection.Specific process is as shown in figure 11: for work order problem new for one, carrying out type of service classification to it, then can lead to
The type of service for manually determining him is crossed, thus, further, the content characteristic class in the content characteristic subset can be modified.
Since the classification of content characteristic has vital reuse for matching efficiency, thus system needs to provide to content
Classification manually adjusts.When business generates change, needs to be adjusted business classification, can neatly be adjusted.
For answer content maintenance: the content of answer is extracted from existing work order solution, then as a mark
Quasi- answer, for solving the problems, such as other similar all work orders.It may need to carry out manual intervention to answer content, make its mark
Standardization and standardization.System provides inquiry and maintenance mechanism for answer.
Threshold value maintenance for answer cluster: in unsupervised clustering algorithm, the answer under same type of service is by being somebody's turn to do
Algorithm carries out the cluster of answer, is directed to similarity threshold miu.Therefore, it can be answered by maintenance threshold value miu to control
Case classification process, if threshold value setting it is higher, more classification can be generated, the work order problematic amount inside each classification compared with
It is few;If threshold value setting is lower, less classification can be generated, internal work order problematic amount is more in each point.
Content characteristic and answer mapping relations are safeguarded: being run by system after criticizing, each answer can all generate it
Corresponding content feature vector, feature vector are the foundations to match as work order question and answer, thus most important.System
System provides maintenance function for feature term vector and answer mapping relations, realizes and constructs new Feature Words in systems to answering
The mapping relations of case, so that accurate match is to answer to the maximum extent.
In conjunction with shown in Figure 12, detailed process is as follows: query feature vector, and the details of feature vector are shown to user, and
It shows answer corresponding to this feature vector, then adjusts feature vector and answer mapping relations, wherein adjustment can be artificial
Interfere feature or machine learning adjustment, judge whether there is manual intervention, if there is manual intervention, is then subject to manual intervention,
If being subject to the result of machine learning without manual intervention.
S330: by work order problem, each of classification answer library corresponding with classification storage work order problem carries out phase one by one
Compare like degree;Classification answer library includes at least a storage work order problem;
Specifically,
Obtain the storage content characteristic of each of classification corresponding classification answer library storage work order;
Based on the cosine law, content characteristic and storage content characteristic are subjected to similarity-rough set.
Specifically, it is previously mentioned, the work order problem of reporting of user is extracted into sign, wherein feature is deposited in the form of vectors
, likewise, for the feature of storage work order problem, is formed and existed with vector, for the space vector of two n dimensions (n > 1),
Its cosine similarity are as follows:
S340: determine that the corresponding storage work order solution to the problem of similarity maximum value to recommend answer, is pushed to use
Family.
Specifically, the value range of cosine value is, closer to 1, indicate that angle closer to 0 degree, that is, two texts
This feature vector is more similar, finds the structure of cosine value closer to 1 as similarity maximum value, then the similarity is maximum
It is worth corresponding storage work order solution to the problem, to recommend answer, is pushed to user.
The worksheet method of the embodiment of the present invention, can be by receiving the counsel requests of reporting of user, in counsel requests
Work order problem is carried, include at least one of the following contents in work order problem: menu code, transaction code, report an error code and work order problem
The problem of describing, determining work order problem type, problem types include first problem type and Second Problem type, first problem class
Type is type the problem of including one of the following contents menu code, transaction code, report an error yard, and Second Problem type is to include at least work order
The problem of problem describes type is based on problem types, obtains recommending answer, is pushed to user, and the present invention is by by work order problem
Classify, and be based on different problems type, obtains push answer, be pushed to user, improve the matching precision of answer, increase
The strong experience sense of user.
In conjunction with shown in Figure 13, a kind of structure chart of worksheet system is shown, the system comprises: receiving unit 510,
Determination unit 520 and push unit 530.
Wherein, receiving unit 510 carry in the counsel requests for the counsel requests for receiving reporting of user
Work order problem, includes at least one of the following contents in the work order problem: menu code, transaction code report an error code and work order problem is retouched
It states;
Determination unit 520, type the problem of for determining the work order problem;Described problem type includes first problem class
Type and Second Problem type;The first problem type is the problem of including one of the following contents menu code, transaction code, report an error yard
Type, the Second Problem type are to include at least the problem of work order problem describes type;
Push unit 530 obtains recommending answer, is pushed to user for being based on problem types.
The worksheet system of the embodiment of the present invention can receive the counsel requests of reporting of user by receiving unit, consult
Ask in request and carry work order problem, include at least one of the following contents in work order problem: menu code, transaction code, report an error code and
The problem of work order problem describes, and determination unit determines work order problem type, problem types include that first problem type and second are asked
Type is inscribed, first problem type is type the problem of including one of the following contents menu code, transaction code, report an error yard, Second Problem
Type is to include at least the problem of work order problem describes type, and push unit is based on problem types, obtains recommending answer, be pushed to
User, the present invention are based on different problems type by the way that work order problem is classified, and obtain push answer, are pushed to use
Family improves the matching precision of answer, enhances the experience sense of user.
Optionally, push unit 530 is specifically used for after determining that described problem type is first problem type, answers from trigram
Whether compare in case library has solution to match with the first problem type;
If it is, extract the solution to match with the first problem type, using the solution as
Answer is recommended to be pushed to the user.
Optionally, push unit 530 is specifically used for after determining that described problem type is Second Problem type, extracts content
Feature;The content characteristic is to extract from the work order problem description in work order problem;
Based on the content characteristic, classification belonging to the work order problem is determined;Each classification is corresponding with classification answer
Library;
By the work order problem one by one each of classification answer library corresponding with classification storage work order problem into
Row similarity-rough set;Classification answer library includes at least a storage work order problem;
Determine that the corresponding storage work order solution to the problem of similarity maximum value to recommend answer, is pushed to the use
Family.
Optionally, the extraction content characteristic in push unit 530, comprising:
Multiple text participles are extracted from the work order problem description in work order problem;
Using TF-IDF, the content characteristic is extracted from the multiple text participle
Optionally, multiple texts are extracted in the work order problem description from work order problem in push unit 530
After the step of participle, it is described utilize TF-IDF, from the multiple text participle in extract the content characteristic the step of it
Before, specifically it is also used to:
To the multiple original text participle be filtered processing, with filter out low-frequency word, stop words, mark information, together
Adopted word.
Optionally, in push unit 530 based on the content characteristic, determine classification belonging to the work order problem, wrap
It includes:
Determine the work order type of service of the work order problem;
According to the work order type of service, reverse classifier is obtained;
The content characteristic is input in the reverse classifier, determines classification belonging to the work order problem.
Optionally, the reverse classifier in push unit 530 obtains by the following method:
The content characteristic of classification and multiple storage work order problems to multiple storage work order problems carries out automatic cluster, obtains
Reverse classifier;
And/or manual command is received, it is based on the manual command, optimizes the reverse classifier.
Optionally, in push unit 530 will be in the work order problem one by one classification answer library corresponding with the classification
Each storage work order problem carry out similarity-rough set, comprising:
Obtain the storage content characteristic of each of classification answer library storage work order problem;
Based on the cosine law, the content characteristic and the storage content characteristic are subjected to similarity-rough set.
The technical effect and preceding method embodiment phase of system provided by the embodiment of the present invention, realization principle and generation
Together, to briefly describe, system embodiment part does not refer to place, can refer to corresponding contents in preceding method embodiment.
The computer program product of worksheet method is carried out provided by the embodiment of the present invention, including stores processor
The computer readable storage medium of executable non-volatile program code, the instruction that said program code includes can be used for executing
Previous methods method as described in the examples, specific implementation can be found in embodiment of the method, and details are not described herein.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description
It with the specific work process of unit, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
In several embodiments provided herein, it should be understood that disclosed system and method can pass through it
Its mode is realized.The apparatus embodiments described above are merely exemplary, for example, the division of the unit, only
A kind of logical function partition, there may be another division manner in actual implementation, in another example, multiple units or components can combine
Or it is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed phase
Coupling, direct-coupling or communication connection between mutually can be through some communication interfaces, the INDIRECT COUPLING of device or unit or
Communication connection can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in the executable non-volatile computer-readable storage medium of a processor.Based on this understanding, of the invention
Technical solution substantially the part of the part that contributes to existing technology or the technical solution can be with software in other words
The form of product embodies, which is stored in a storage medium, including some instructions use so that
One computer equipment (can be personal computer, server or the network equipment etc.) executes each embodiment institute of the present invention
State all or part of the steps of method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-
Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can be with
Store the medium of program code.
The embodiment of the invention also provides a kind of electronic equipment, including memory, processor, it is stored in the memory
The computer program that can be run on the processor, the processor realize above-described embodiment when executing the computer program
The step of described in any item methods.
Wherein, memory may include high-speed random access memory (RAM, Random Access Memory), can also
It can further include non-labile memory (non-volatile memory), for example, at least a magnetic disk storage.
Processor may be a kind of IC chip, the processing capacity with signal.During realization, the above method
Each step can be completed by the instruction of the integrated logic circuit of the hardware in processor or software form.Above-mentioned processing
Device can be general processor, including central processing unit (Central Processing Unit, abbreviation CPU), network processes
Device (Network Processor, abbreviation NP) etc.;It can also be digital signal processor (Digital Signal
Processing, abbreviation DSP), specific integrated circuit (Application Specific Integrated Circuit, referred to as
ASIC), ready-made programmable gate array (Field-Programmable Gate Array, abbreviation FPGA) or other are programmable
Logical device, discrete gate or transistor logic, discrete hardware components.It may be implemented or execute in the embodiment of the present invention
Disclosed each method, step and logic diagram.General processor can be microprocessor or the processor is also possible to appoint
What conventional processor etc..The step of method in conjunction with disclosed in the embodiment of the present invention, can be embodied directly in hardware decoding processing
Device executes completion, or in decoding processor hardware and software module combination execute completion.Software module can be located at
Machine memory, flash memory, read-only memory, programmable read only memory or electrically erasable programmable memory, register etc. are originally
In the storage medium of field maturation.The storage medium is located at memory, and processor reads the information in memory, in conjunction with its hardware
The step of completing the above method.
Unless specifically stated otherwise, the opposite step of the component and step that otherwise illustrate in these embodiments, digital table
It is not limit the scope of the invention up to formula and numerical value.
In all examples being illustrated and described herein, any occurrence should be construed as merely illustratively, without
It is as limitation, therefore, other examples of exemplary embodiment can have different values.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.
The flow chart and block diagram in the drawings show the system of multiple embodiments according to the present invention, method and computer journeys
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, section or code of table, a part of the module, section or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two continuous boxes can actually base
Originally it is performed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.It is also noted that
It is the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, execution can be used
Defined function or the dedicated hardware based system of movement realize, or can use specialized hardware and computer instruction
Combination is to realize.
In addition, in the description of the embodiment of the present invention unless specifically defined or limited otherwise, term " installation ", " phase
Even ", " connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected;It can
To be mechanical connection, it is also possible to be electrically connected;It can be directly connected, can also can be indirectly connected through an intermediary
Connection inside two elements.For the ordinary skill in the art, above-mentioned term can be understood at this with concrete condition
Concrete meaning in invention.
In the description of the present invention, it should be noted that term " center ", "upper", "lower", "left", "right", "vertical",
The orientation or positional relationship of the instructions such as "horizontal", "inner", "outside" be based on the orientation or positional relationship shown in the drawings, merely to
Convenient for description the present invention and simplify description, rather than the device or element of indication or suggestion meaning must have a particular orientation,
It is constructed and operated in a specific orientation, therefore is not considered as limiting the invention.In addition, term " first ", " second ",
" third " is used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance.
Finally, it should be noted that embodiment described above, only a specific embodiment of the invention, to illustrate the present invention
Technical solution, rather than its limitations, scope of protection of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair
It is bright to be described in detail, those skilled in the art should understand that: anyone skilled in the art
In the technical scope disclosed by the present invention, it can still modify to technical solution documented by previous embodiment or can be light
It is readily conceivable that variation or equivalent replacement of some of the technical features;And these modifications, variation or replacement, do not make
The essence of corresponding technical solution is detached from the spirit and scope of technical solution of the embodiment of the present invention, should all cover in protection of the invention
Within the scope of.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of worksheet method characterized by comprising
The counsel requests of reporting of user are received, work order problem is carried in the counsel requests, is at least wrapped in the work order problem
Include one of the following contents: menu code, transaction code, report an error code and the description of work order problem;
The problem of determining work order problem type;Described problem type includes first problem type and Second Problem type;Institute
Stating first problem type is type the problem of including at least one of the following contents menu code, transaction code, report an error yard, and described second asks
Topic type is to include at least the problem of work order problem describes type;
Based on described problem type, obtains recommending answer, be pushed to the user.
2. obtaining recommending answer, being pushed to the method according to claim 1, wherein being based on described problem type
The user, comprising:
Determine described problem type be first problem type after, compared from trigram answer library whether have solution with it is described
First problem type matches;
If it is, the solution to match with the first problem type is extracted, using the solution as recommendation
Answer is pushed to the user.
3. obtaining recommending answer, being pushed to the method according to claim 1, wherein being based on described problem type
The user, comprising:
After determining that described problem type is Second Problem type, content characteristic is extracted;The content characteristic is from work order problem
In the description of work order problem in extract;
Based on the content characteristic, classification belonging to the work order problem is determined;Each classification is corresponding with classification answer library;
By the work order problem, each of classification answer library corresponding with classification storage work order problem carries out phase one by one
Compare like degree;Classification answer library includes at least a storage work order problem;
Determine that the corresponding storage work order solution to the problem of similarity maximum value to recommend answer, is pushed to the user.
4. according to the method described in claim 3, it is characterized in that, extracting content characteristic, comprising:
Multiple text participles are extracted from the work order problem description in work order problem;
Using TF-IDF, the content characteristic is extracted from the multiple text participle.
5. according to the method described in claim 4, it is characterized in that, being mentioned in the work order problem description from work order problem
After the step of taking multiple texts to segment, TF-IDF is utilized described, it is special to extract the content from the multiple text participle
Before the step of sign, the method also includes:
Processing is filtered to the multiple original text participle, to filter out low-frequency word, stop words, mark information, synonym.
6. according to the method described in claim 3, it is characterized in that, determining the work order problem institute based on the content characteristic
The classification of category, comprising:
Determine the work order type of service of the work order problem;
According to the work order type of service, reverse classifier is obtained;
The content characteristic is input in the reverse classifier, determines classification belonging to the work order problem.
7. according to the method described in claim 6, it is characterized in that, the reverse classifier obtains by the following method:
The content characteristic of classification and multiple storage work order problems to multiple storage work order problems carries out automatic cluster, obtains reverse
Classifier;
And/or manual command is received, it is based on the manual command, optimizes the reverse classifier.
8. according to the method described in claim 3, it is characterized in that, by work order problem class corresponding with the classification one by one
Each of other answer library storage work order problem carries out similarity-rough set, comprising:
Obtain the storage content characteristic of each of classification answer library storage work order problem;
Based on the cosine law, the content characteristic and the storage content characteristic are subjected to similarity-rough set.
9. a kind of worksheet system characterized by comprising
Receiving unit carries work order problem, the work order for receiving the counsel requests of reporting of user in the counsel requests
One of the following contents is included at least in problem: menu code, transaction code, report an error code and the description of work order problem;
Determination unit, type the problem of for determining the work order problem;Described problem type includes first problem type and
Two problem types;The first problem type is type the problem of including one of the following contents menu code, transaction code, report an error yard,
The Second Problem type is to include at least the problem of work order problem describes type;
Push unit obtains recommending answer, is pushed to the user for being based on described problem type.
10. a kind of electronic equipment, including memory, processor, it is stored with and can runs on the processor in the memory
Computer program, which is characterized in that the processor realizes the claims 1 to 8 when executing the computer program
The step of method described in one.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810682171.5A CN109344227A (en) | 2018-06-27 | 2018-06-27 | Worksheet method, system and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810682171.5A CN109344227A (en) | 2018-06-27 | 2018-06-27 | Worksheet method, system and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109344227A true CN109344227A (en) | 2019-02-15 |
Family
ID=65296735
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810682171.5A Pending CN109344227A (en) | 2018-06-27 | 2018-06-27 | Worksheet method, system and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109344227A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109885768A (en) * | 2019-02-18 | 2019-06-14 | 中国联合网络通信集团有限公司 | Worksheet method, apparatus and system |
CN110442873A (en) * | 2019-08-07 | 2019-11-12 | 云南电网有限责任公司信息中心 | A kind of hot spot work order acquisition methods and device based on CBOW model |
CN110502622A (en) * | 2019-07-03 | 2019-11-26 | 平安科技(深圳)有限公司 | Common medical question and answer data creation method, device and computer equipment |
CN110737698A (en) * | 2019-10-15 | 2020-01-31 | 重庆浪尖至简物联网科技有限公司 | question-related information recommendation method based on question description |
CN112214508A (en) * | 2020-10-20 | 2021-01-12 | 政采云有限公司 | Data processing method and device |
CN117009605A (en) * | 2023-08-08 | 2023-11-07 | 四川大学 | Strategic innovation design problem solving method and system |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105610609A (en) * | 2015-12-17 | 2016-05-25 | 美的集团股份有限公司 | Method and system for processing fault of household electrical appliance, household electrical appliance and cloud server |
CN106506645A (en) * | 2016-11-10 | 2017-03-15 | 中车青岛四方机车车辆股份有限公司 | The monitoring method and system of rail vehicle |
CN107168285A (en) * | 2017-05-26 | 2017-09-15 | 大连理工大学 | A kind of automobile intelligent fault diagnosis of combination subjective and objective information and cloud model and maintenance householder method and system |
CN107161821A (en) * | 2017-05-25 | 2017-09-15 | 梯联网(贵州)科技股份有限公司 | A kind of elevator terminal fault resolution |
WO2017219317A1 (en) * | 2016-06-23 | 2017-12-28 | 北京三快在线科技有限公司 | Information pushing method and device based on search content |
CN108053351A (en) * | 2018-02-08 | 2018-05-18 | 南京邮电大学 | Intelligent college entrance will commending system and recommendation method |
-
2018
- 2018-06-27 CN CN201810682171.5A patent/CN109344227A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105610609A (en) * | 2015-12-17 | 2016-05-25 | 美的集团股份有限公司 | Method and system for processing fault of household electrical appliance, household electrical appliance and cloud server |
WO2017219317A1 (en) * | 2016-06-23 | 2017-12-28 | 北京三快在线科技有限公司 | Information pushing method and device based on search content |
CN106506645A (en) * | 2016-11-10 | 2017-03-15 | 中车青岛四方机车车辆股份有限公司 | The monitoring method and system of rail vehicle |
CN107161821A (en) * | 2017-05-25 | 2017-09-15 | 梯联网(贵州)科技股份有限公司 | A kind of elevator terminal fault resolution |
CN107168285A (en) * | 2017-05-26 | 2017-09-15 | 大连理工大学 | A kind of automobile intelligent fault diagnosis of combination subjective and objective information and cloud model and maintenance householder method and system |
CN108053351A (en) * | 2018-02-08 | 2018-05-18 | 南京邮电大学 | Intelligent college entrance will commending system and recommendation method |
Non-Patent Citations (1)
Title |
---|
李晓鹏: "《计算机网络》课程FAQ系统设计与开发", 《中国优秀硕士学位论文全文数据库社会科学辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109885768A (en) * | 2019-02-18 | 2019-06-14 | 中国联合网络通信集团有限公司 | Worksheet method, apparatus and system |
CN110502622A (en) * | 2019-07-03 | 2019-11-26 | 平安科技(深圳)有限公司 | Common medical question and answer data creation method, device and computer equipment |
CN110442873A (en) * | 2019-08-07 | 2019-11-12 | 云南电网有限责任公司信息中心 | A kind of hot spot work order acquisition methods and device based on CBOW model |
CN110737698A (en) * | 2019-10-15 | 2020-01-31 | 重庆浪尖至简物联网科技有限公司 | question-related information recommendation method based on question description |
CN112214508A (en) * | 2020-10-20 | 2021-01-12 | 政采云有限公司 | Data processing method and device |
CN117009605A (en) * | 2023-08-08 | 2023-11-07 | 四川大学 | Strategic innovation design problem solving method and system |
CN117009605B (en) * | 2023-08-08 | 2024-04-02 | 四川大学 | Strategic innovation design problem solving method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116628172B (en) | Dialogue method for multi-strategy fusion in government service field based on knowledge graph | |
CN112612902B (en) | Knowledge graph construction method and device for power grid main equipment | |
CN109344227A (en) | Worksheet method, system and electronic equipment | |
CN109063221A (en) | Query intention recognition methods and device based on mixed strategy | |
CN110659282B (en) | Data route construction method, device, computer equipment and storage medium | |
CN108416032A (en) | A kind of file classification method, device and storage medium | |
KR101953190B1 (en) | A multidimensional recursive learning process and system used to discover complex dyadic or multiple counterparty relationships | |
CN115547466B (en) | Medical institution registration and review system and method based on big data | |
CN110046264A (en) | A kind of automatic classification method towards mobile phone document | |
CN116089873A (en) | Model training method, data classification and classification method, device, equipment and medium | |
CN112632982A (en) | Dialogue text emotion analysis method capable of being used for supplier evaluation | |
CN115146062A (en) | Intelligent event analysis method and system fusing expert recommendation and text clustering | |
Ma et al. | Content Feature Extraction-based Hybrid Recommendation for Mobile Application Services. | |
CN114491079A (en) | Knowledge graph construction and query method, device, equipment and medium | |
CN116737758A (en) | Database query statement generation method, device, equipment and storage medium | |
CN115239214B (en) | Enterprise evaluation processing method and device and electronic equipment | |
CN106775694A (en) | A kind of hierarchy classification method of software merit rating code product | |
CN110413757A (en) | A kind of word paraphrase determines method, apparatus and system | |
CN113177164B (en) | Multi-platform collaborative new media content monitoring and management system based on big data | |
US20210103593A1 (en) | Model validation for query intent | |
AU2019290658B2 (en) | Systems and methods for identifying and linking events in structured proceedings | |
CN114610882A (en) | Abnormal equipment code detection method and system based on electric power short text classification | |
CN114048325A (en) | Case processing method and device based on knowledge graph and storage medium | |
CN114254622A (en) | Intention identification method and device | |
CN112905747A (en) | Professional system archive question-answering robot system based on semantic analysis technology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190215 |