CN110930022A - Hotel static information detection method and system, electronic equipment and storage medium - Google Patents
Hotel static information detection method and system, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110930022A CN110930022A CN201911142094.5A CN201911142094A CN110930022A CN 110930022 A CN110930022 A CN 110930022A CN 201911142094 A CN201911142094 A CN 201911142094A CN 110930022 A CN110930022 A CN 110930022A
- Authority
- CN
- China
- Prior art keywords
- hotel
- information
- feedback information
- static information
- static
- 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
- 230000003068 static effect Effects 0.000 title claims abstract description 275
- 238000001514 detection method Methods 0.000 title claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 64
- 238000010801 machine learning Methods 0.000 claims abstract description 32
- 238000012360 testing method Methods 0.000 claims description 46
- 238000011156 evaluation Methods 0.000 claims description 30
- 238000006243 chemical reaction Methods 0.000 claims description 20
- 238000012937 correction Methods 0.000 claims description 19
- 238000007781 pre-processing Methods 0.000 claims description 14
- 238000013136 deep learning model Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 10
- 238000000034 method Methods 0.000 abstract description 32
- 239000013598 vector Substances 0.000 description 22
- 230000009182 swimming Effects 0.000 description 18
- 238000013135 deep learning Methods 0.000 description 12
- 238000010276 construction Methods 0.000 description 5
- 238000007635 classification algorithm Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
- G06Q30/012—Providing warranty services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/12—Hotels or restaurants
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Finance (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Accounting & Taxation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention discloses a method and a system for detecting hotel static information, electronic equipment and a storage medium, wherein the detection method comprises the following steps: acquiring first feedback information of a user on the hotel, wherein the first feedback information comprises comment information of the user on the hotel; inputting the first feedback information into a static information prediction model to obtain second feedback information comprising hotel static information, wherein the static information prediction model is obtained after training a machine learning model based on historical feedback information, and the historical feedback information comprises historical comment information of a user on the hotel; acquiring hotel static information provided by a hotel; and judging whether the content of the hotel static information in the second feedback information is consistent with the content of the hotel static information provided by the hotel, and if not, checking the inconsistent static information. The method and the system can correct wrong static information provided by the hotel in time, improve the accuracy of the static information of the hotel, improve the check-in experience of the user, reduce the complaint amount of the user and improve the OTA brand image.
Description
Technical Field
The invention relates to the technical field of computer information, in particular to a hotel static information detection method and system, electronic equipment and a storage medium.
Background
In OTA (on-line travel agency) industry, travel consumers reserve travel products or services to travel service providers through a network and pay online or offline, namely, each travel subject can carry out product marketing or product selling through the network, the original traditional travel agency selling mode is put on a network platform, line information is transmitted more widely, and interactive communication is more convenient for customers to consult and order.
In the current OTA industry, the verification of the accuracy of the static information provided by the hotel is mainly performed by manually verifying the telephone of the hotel, when the number of hotels is large, the verification method needs to consume a large amount of human resources, the static information of the hotel can change at any time, and needs to be manually verified again, so that a large amount of human resource cost is required repeatedly.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a detection method, a detection system, an electronic device and a storage medium for hotel static information, aiming at overcoming the defects that in the prior art, static information provided by a hotel cannot be updated in time, so that the accuracy of the static information is difficult to verify, and misleading is caused to a user, and thus the user experience is poor.
The invention solves the technical problems through the following technical scheme:
the invention provides a method for detecting hotel static information, which comprises the following steps:
acquiring first feedback information of a user on a hotel, wherein the first feedback information comprises comment information of the user on the hotel;
inputting the first feedback information into a static information prediction model to obtain second feedback information comprising hotel static information, wherein the static information prediction model is obtained after training a machine learning model based on historical feedback information, and the historical feedback information comprises historical comment information of a user on the hotel;
acquiring hotel static information provided by the hotel;
and judging whether the content of the hotel static information in the second feedback information is consistent with the content of the corresponding hotel static information provided by the hotel, and if not, checking the inconsistent static information.
The deep learning text classification method in the machine learning model needs to collect text corpora in advance, construct Word vectors of a text, and then use a deep learning classification algorithm Word vector based on the existing Word vectors, wherein the construction method of the Word vectors includes Word2Vec (a Word vector construction method) and GloVe (a Word vector construction method).
According to the method and the device, the comment condition of the user on the hotel is fully utilized, the possible problems in the static hotel information provided by the hotel are timely excavated, whether the static hotel information provided by the hotel is correct or not is judged by comparing the static hotel information fed back by the user with the static hotel information provided by the hotel, and if the static hotel information provided by the hotel is incorrect, error correction is timely carried out, so that the accuracy of the static hotel information is improved.
Preferably, the step of verifying said static information as inconsistent comprises:
judging whether the quantity of second feedback information inconsistent with the corresponding hotel static information provided by the hotel exceeds an error correction information threshold value, and if so, verifying the corresponding hotel static information provided by the hotel;
and/or the presence of a gas in the gas,
the first feedback information also comprises complaint information of the user to the hotel and dialogue information of the user and customer service.
According to the invention, by setting the error correction threshold, the influence of malicious comment on the detection method can be avoided, so that the accuracy of the detection of the hotel static information provided by the hotel is improved.
According to the invention, the first feedback information is acquired through various channels, so that problematic static information provided by the hotel can be detected more comprehensively and accurately.
Preferably, the detection method further comprises:
obtaining the static information prediction model by:
acquiring historical feedback information of a user on a hotel;
selecting historical feedback information with the quantity of a preset numerical value;
judging whether the selected historical feedback information comprises the hotel static information, if so, marking the historical feedback information as a first machine code, and if not, marking the historical feedback information as a second machine code;
dividing the marked historical feedback information into a training set and a test set;
taking the historical feedback information in the training set as input and taking the first machine code and the second machine code which are correspondingly marked as output, and training the machine codes in a machine learning model;
inputting the historical feedback information in the test set into a trained machine learning model to obtain a test machine code corresponding to each piece of historical feedback information;
and judging whether the evaluation index AUC of the test machine code reaches an evaluation threshold value, if so, the trained machine learning model is the static information prediction model.
The evaluation index AUC is an area under an ROC curve, an abscissa of the area under the ROC curve is a ratio of the number of second machine codes to the number of test machine codes inconsistent with the number of corresponding marked machine codes, where the test machine code is the first machine code, and the ordinate of the area under the ROC curve is a ratio of the number of first machine codes to the number of first machine codes consistent with the number of first machine codes and corresponding marked machine codes.
According to the static information prediction method and device, the historical feedback information is input into the machine model to be trained to obtain the static information prediction model, so that whether the static information exists in the feedback information can be screened by the model method, the workload of manually distinguishing whether each piece of the feedback information has the static information or not can be greatly reduced, the labor force is greatly saved, and the detection efficiency is improved.
According to the method, the condition that the static information is inaccurate due to inaccurate model training can be avoided through the set AUC (area under the curve) index, and a more accurate static information prediction model can be further trained.
Preferably, if the evaluation index AUC is judged not to reach the evaluation threshold, adding the labeled new historical feedback information into the training set for training again;
and/or the presence of a gas in the gas,
the step of selecting the historical feedback information with the quantity of the preset numerical value further comprises the following steps: preprocessing the historical feedback information, wherein the preprocessing comprises at least one of sensitive word removal, special character removal, full angle conversion of half angle, traditional Chinese character conversion to simple Chinese character conversion and upper case conversion to lower case conversion;
and/or the presence of a gas in the gas,
the historical feedback information is historical comment information with the comment score lower than a comment score threshold value.
In the invention, when the AUC index does not reach the threshold value, the model can be updated in real time by retraining the model so as to obtain a more accurate static information prediction model.
According to the method and the device, the historical comment information lower than the point score threshold value is trained, and the static information provided by the hotel and inconsistent with the real situation can be acquired more pertinently according to the negative feedback of the user, so that the static information of the relevant hotel can be verified conveniently, and the accuracy of the static information provided by the hotel is improved.
Preferably, the step of verifying said static information as inconsistent comprises: judging whether the hotel static information provided by the hotel is consistent with the real static information of the hotel, if not, modifying the hotel static information provided by the hotel into the real static information of the hotel;
and/or the presence of a gas in the gas,
the machine training model is a deep learning model, wherein the deep learning model can be TextCNN (a deep learning model), LSTM (a deep learning model), BI-LSTM (a deep learning model), and the like.
According to the invention, inconsistent static information provided by the hotel can be updated in time, so that the accuracy of the static information of the hotel is improved, the check-in experience of the user is improved, the complaint amount of the user to the hotel is reduced, and the brand image of the OTA is improved.
The invention also provides a detection system of hotel static information, which comprises:
the system comprises a first feedback information acquisition module, a second feedback information acquisition module, a hotel static information acquisition module and a static information judgment module;
the first feedback information acquisition module is used for acquiring first feedback information of a user on the hotel, and the first feedback information comprises comment information of the user on the hotel;
the second feedback information acquisition module is used for inputting the first feedback information into a static information prediction model to acquire second feedback information comprising hotel static information, the static information prediction model is obtained after a machine learning model is trained on the basis of historical feedback information, and the historical feedback information comprises historical comment information of a user on the hotel;
the hotel static information acquisition module is used for acquiring hotel static information provided by the hotel;
the static information judgment module is used for judging whether the content of the hotel static information in the second feedback information is consistent with the content of the corresponding hotel static information provided by the hotel, and if not, the inconsistent static information is verified.
The deep learning text classification method in the machine learning model needs to collect text corpora in advance, construct Word vectors of a text, and then use a deep learning classification algorithm Word vector based on the existing Word vectors, wherein the construction method of the Word vectors includes Word2Vec and GloVe.
According to the method and the device, the comment condition of the user on the hotel is fully utilized through the first feedback information acquisition module, the possible problems in the static information of the hotel provided by the hotel are timely excavated through the second feedback information acquisition module, the static information of the hotel fed back by the user is compared with the static information of the hotel provided by the hotel through the static information judgment module, so that whether the static information of the hotel provided by the hotel is correct or not is judged, and if the static information of the hotel provided by the hotel is incorrect, error correction is timely carried out, so that the accuracy of the static information of the hotel is improved.
Preferably, the static information determining module is further configured to determine whether the number of second feedback information inconsistent with the corresponding hotel static information provided by the hotel exceeds an error correction information threshold, and if so, verify the corresponding hotel static information provided by the hotel;
and/or the presence of a gas in the gas,
the first feedback information also comprises complaint information of the user to the hotel and dialogue information of the user and customer service.
According to the invention, through setting the error correction threshold value in the static information judgment module, the influence of malicious comment on the detection system can be avoided, so that the accuracy of the detection of the hotel static information provided by the hotel is improved.
According to the invention, the first feedback information is acquired through various channels, so that problematic static information provided by the hotel can be detected more comprehensively and accurately.
Preferably, the detection system further comprises:
the model obtaining module is used for obtaining the static information prediction model;
the model acquisition module comprises: the device comprises a historical feedback information acquisition unit, a historical information selection unit, a selection information judgment unit, a division unit, a training unit, a test unit and an index judgment unit;
the historical feedback information acquisition unit is used for acquiring historical feedback information of a user on the hotel and calling the historical information selection unit;
the history information selection unit is used for selecting history feedback information with the number of preset values and calling the selected information judgment unit;
the selected information judging unit is used for judging whether the selected historical feedback information comprises the hotel static information, if so, the historical feedback information is marked as a first machine code, if not, the historical feedback information is marked as a second machine code, and the dividing unit is called;
the dividing unit is used for dividing the marked historical feedback information into a training set and a test set and calling the training unit;
the training unit is used for taking the historical feedback information in the training set as input and taking the first machine code and the second machine code which are correspondingly marked as output, training the historical feedback information in a machine learning model, and calling the testing unit;
the test unit is used for inputting the historical feedback information in the test set into a trained machine learning model to obtain a test machine code corresponding to each piece of historical feedback information, and calling the index judgment unit;
the index judgment unit is used for judging whether the evaluation index AUC of the test machine code reaches an evaluation threshold value, and if the evaluation index AUC reaches the evaluation threshold value, the trained machine learning model is the static information prediction model.
The evaluation index AUC is an area under an ROC (change rate index) curve, an abscissa of the area under the ROC curve is a ratio of the number of the second machine codes to the number of the test machine codes and the number of the machine codes of the corresponding marks to be inconsistent with the number of the test machine codes and the machine codes of the corresponding marks, and an ordinate of the area under the ROC curve is a ratio of the number of the first machine codes to the number of the machine codes of the first machine codes to be consistent with the number of the machine codes of the first machine codes and the corresponding marks to be consistent with each other.
According to the static information prediction method and device, the historical feedback information is input into the machine model through the model acquisition module to be trained to obtain the static information prediction model, so that whether the static information exists in the feedback information can be screened through a model method, the workload of manually distinguishing whether each piece of a large amount of feedback information exists in the static information can be greatly reduced, the labor force is greatly saved, and the detection efficiency is improved.
According to the invention, the condition that the model training is inaccurate and the predicted static information is inaccurate can be avoided through the AUC index set by the index judgment unit, and a more accurate static information prediction model can be further trained.
Preferably, if the index judgment unit judges that the evaluation index AUC does not reach the evaluation threshold, the index judgment unit invokes the training unit, and the training unit is further configured to add the labeled new historical feedback information into the training set for training again;
and/or the presence of a gas in the gas,
the model acquisition module further comprises: the preprocessing unit is used for preprocessing the historical feedback information and comprises at least one of sensitive word removal, special character removal, full angle conversion of half angle, traditional Chinese character conversion to simplified Chinese character conversion and upper case conversion to lower case conversion;
and/or the presence of a gas in the gas,
the historical feedback information is historical comment information with the comment score lower than a comment score threshold value.
In the invention, when the judgment unit judges that the AUC index does not reach the threshold value, the training unit can update the model in real time by retraining the model so as to obtain a more accurate static information prediction model.
According to the method and the device, the training unit is used for training the historical comment information which is lower than the point score threshold value, and the static information provided by the hotel and inconsistent with the real situation can be acquired according to the negative feedback of the user, so that the static information of the relevant hotel can be verified conveniently, and the accuracy of the static information provided by the hotel is improved.
Preferably, the static information determining module is further configured to determine whether hotel static information provided by the hotel is consistent with the real static information of the hotel, and if not, modify the hotel static information provided by the hotel into the real static information of the hotel;
and/or the presence of a gas in the gas,
the machine training model is a deep learning model, wherein the deep learning model can be TextCNN, LSTM, BI-LSTM, and the like.
According to the invention, the static information judgment module can update inconsistent static information provided by the hotel in time, so that the accuracy of the static information of the hotel is improved, the check-in experience of a user is improved, the complaint amount of the user to the hotel is reduced, and the brand image of the OTA is improved.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the detection method when executing the computer program.
The invention also provides a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the detection method described above.
The positive progress effects of the invention are as follows: according to the method and the device, the comment condition of the user on the hotel is fully utilized, the possible problems in the hotel static information provided by the hotel are timely excavated, whether the hotel static information provided by the hotel is correct or not is judged by comparing the hotel static information fed back by the user with the hotel static information provided by the hotel, and if the hotel static information provided by the hotel is incorrect, error correction is timely carried out, so that the accuracy of the hotel static information is improved, the check-in experience of the user is improved, the complaint amount of the user is reduced, and the brand OTA image is improved.
Drawings
Fig. 1 is a flowchart of a method for detecting hotel static information according to embodiment 1 of the present invention
Fig. 2 is a partial flowchart of a method for detecting hotel static information according to embodiment 2 of the present invention.
Fig. 3 is a flowchart of the steps of obtaining a static information prediction model in embodiment 3 of the present invention.
Fig. 4 is a schematic block diagram of a system for detecting hotel static information according to embodiment 5 of the present invention
Fig. 5 is a module schematic view of a model obtaining module in the hotel static information detection system according to embodiment 7 of the present invention.
Fig. 6 is a schematic diagram of a hardware structure of an electronic device according to embodiment 9 of the present invention.
Detailed Description
The invention is further illustrated by the following examples, which are not intended to limit the scope of the invention.
Example 1
The embodiment provides a method for detecting hotel static information, as shown in fig. 1, the method includes:
And 102, inputting the first feedback information into a static information prediction model to obtain second feedback information comprising hotel static information.
And step 103, obtaining hotel static information provided by the hotel.
And 104, judging whether the content of the hotel static information in the second feedback information is consistent with the content of the corresponding hotel static information provided by the hotel, if so, executing a step 105, and if not, executing a step 106.
And 105, confirming that the corresponding hotel static information provided by the hotel is correct.
And step 106, verifying the inconsistent static information.
The first feedback information acquired in step 101 includes comment information of the user on the hotel.
The static information prediction model in the step 102 is obtained by training a machine learning model based on historical feedback information, the historical feedback information includes historical comment information of a user on a hotel, the deep learning text classification method in the machine learning model needs to collect text corpora in advance to construct Word vectors of a text, and then deep learning classification algorithm Word vectors are used based on existing Word vectors, and the construction method of the Word vectors is Word2Vec or GloVe.
According to the embodiment, the comment condition of the user on the hotel is fully utilized, the possible problems in the hotel static information provided by the hotel are timely excavated, the hotel static information fed back by the user is compared with the hotel static information provided by the hotel, so that whether the hotel static information provided by the hotel is correct or not is judged, if not, error correction is timely carried out, and the accuracy of the hotel static information is improved.
Example 2
This embodiment is a method for detecting hotel static information, which is further improved on the basis of embodiment 1, as shown in fig. 2, this embodiment is basically the same as the steps of embodiment 1, except that, in step 104, if it is determined that the content of the hotel static information in the second feedback information is not consistent with the content of the corresponding hotel static information provided by the hotel, step 201 is executed.
And step 203, modifying the hotel static information provided by the hotel into real static information of the hotel.
And step 204, confirming that the corresponding hotel static information provided by the hotel is correct.
In the embodiment, the error correction threshold is set, so that the influence of malicious comment on the detection method can be avoided, and the accuracy of the detection of the hotel static information provided by the hotel is improved.
In this embodiment, inconsistent static information that the hotel provided can in time be updated to promote the accuracy of the static information in hotel, promote user's experience of living in, reduce the amount of complaints of user to the hotel, promote OTA's brand image.
Example 3
This example is a further modification of example 1 or example 2, and in this example,
the first feedback information also comprises complaint information of the user to the hotel, dialogue information of the user and customer service and other related information which can be obtained and fed back to the hotel by the user.
As shown in fig. 3, the static information prediction model in the present embodiment is obtained through the following steps:
And step 303, preprocessing historical feedback information.
And step 306, marking the historical feedback information as a second machine code.
And 307, dividing the marked historical feedback information into a training set and a test set.
And 308, taking the historical feedback information in the training set as input and the first machine code and the second machine code which are correspondingly marked as output, and training the machine codes in the machine learning model.
And 310, judging whether the evaluation index AUC of the test machine code reaches an evaluation threshold value, if so, executing step 311, and if not, executing step 312.
In step 303, the preprocessing may be to remove sensitive words, remove special characters, change the half angle from full angle, change from traditional to simple, change from upper case to lower case, and so on.
In step 308, the machine learning model may be a deep learning model that may be obtained by TextCNN, LSTM, BI-LSTM, and the like.
The evaluation index AUC in step 310 is an area under an ROC curve, an abscissa of the area under the ROC curve is a ratio of the number of the second machine codes to the number of the test machine codes and the number of the machine codes of the corresponding marks that are inconsistent with the number of the test machine codes and the machine codes of the corresponding marks, and an ordinate of the area under the ROC curve is a ratio of the number of the first machine codes to the number of the machine codes of the first machine codes and the corresponding marks that are consistent with the number of the first machine codes and the machine codes of the corresponding marks.
For better understanding of the present embodiment, the present embodiment will be further explained by taking a specific example as follows:
firstly, acquiring comment content of a user on a hotel in the OTA industry, conversation text content of the user and hotel customer service, and common open-source text corpora such as some common newspapers and web news, constructing word vectors for a deep learning text classification method, and constructing the dimension of each obtained word vector to be 100-dimensional or 200-dimensional. And then, extracting historical feedback information of the user on the hotel, and then carrying out pretreatment such as special character removal, full angle turning to half angle, traditional turning to simplified form, upper writing to lower writing and the like on the historical feedback information. And then, selecting a part of the processed historical feedback information for manual marking, and judging whether the processed historical feedback information has the content of static information, if so, marking the processed historical feedback information as 1, otherwise, marking the processed historical feedback information as 0, and if the processed historical feedback information is: when the hotel environment is poor, the piece of content does not include static information, and is marked as 0, and when the feedback information is: the hotel has a parking lot, and since the parking lot is static information, the label of the piece of feedback information is 0. Then, the marked content is randomly divided into a training set A and a testing set B according to a certain proportion (such as 7: 3). Deep learning model training is carried out on historical feedback information in the training set A, the model is verified by using data in the test set B, when the evaluation index AUC of the static information problems in the verified test set reaches a threshold value, the training is stopped to obtain the model, otherwise, the model is continuously trained, wherein the higher the AUC index is, the better the model effect is represented. After the model is trained, the first feedback information is input into the trained model, and in this embodiment, the first feedback information includes 5 pieces of content: the hotel has no delivery service, the hotel environment is elegant, the swimming pool of the hotel is not wrong, the attitude of the customer service staff is poor, the hotel does not provide the swimming pool to obtain the content with the static information problem, and the obtained content is the content with the static information, in the embodiment, the following three contents are obtained: the hotel has no pick-up service, the swimming pool of the hotel is good, and the hotel does not provide the swimming pool. Next, whether feedback information inconsistent with the content of the static information provided by the hotel reaches an error correction information threshold value is determined, in this embodiment, the set error correction threshold value is 5, and only 2 pieces of information of the hotel with no delivery service are provided as the feedback information, but 12 pieces of information of the hotel with no swimming pool are provided as the feedback information, the hotel with delivery service is temporarily defaulted, and for the feedback of the hotel with no swimming pool, the content of the static information provided by the hotel is checked, and the static information provided by the hotel is checked as follows: the hotel has a swimming pool, the feedback information provides no swimming pool for the hotel, the information is inconsistent, then the next step is carried out, whether the static information of the hotel provided by the pitted real hotel is consistent with the real static information of the hotel, and if the real situation of the hotel is that no swimming pool exists, the static information of the hotel can be corrected.
In the embodiment, the historical feedback information is input into the machine model to be trained to obtain the static information prediction model, so that whether the static information exists in the feedback information can be screened by the model method, the workload of manually distinguishing whether each piece of the large amount of feedback information has the static information can be greatly reduced, the labor force is greatly saved, and the detection efficiency is improved.
In this embodiment, by the set AUC indexes, the situation that the model training is not accurate so that the predicted static information is not accurate can be avoided, and a more accurate static information prediction model can be further trained.
In this embodiment, when the AUC index does not reach the threshold, the model may be updated in real time by retraining the model, so as to obtain a more accurate static information prediction model.
In the embodiment, the feedback information of the hotel is fully utilized by utilizing a classification method of deep learning, the feedback information is predicted through the model, the content reflecting the problem of the static information is obtained, the static information item with the problem is extracted to be matched and compared with the static information item provided by the hotel, inconsistent content is mined to be manually verified, the wrong static information content in the hotel is corrected in time, the comment prediction is carried out by using the model method, and the manual judgment of comment texts can be greatly reduced. Meanwhile, through manual verification of the hotel with problems, the cost for manually verifying all hotels once is greatly reduced, the accuracy of static information of the hotel is improved, the check-in experience of a user is improved, the complaint amount of the user to the hotel is reduced, and the brand image of the OTA is improved.
Example 4
This embodiment is basically the same as embodiment 3 except that the historical feedback information in embodiment 3 is replaced with historical point score information having a point score lower than a point score threshold.
The point score threshold value can be a median of a hotel point score range, for example, if the value of the hotel point score is 1-5 points, the threshold value is 3 points.
Because the historical comment information with the point score lower than the point score threshold generally reflects that the content of the static information item of the hotel is problematic, in the embodiment, the static information provided by the hotel, which is inconsistent with the real situation, can be acquired more pertinently according to the negative feedback of the user by training the historical comment information with the point score lower than the point score threshold, so that the relevant hotel can verify the static information conveniently, and the accuracy of the static information provided by the hotel is improved.
Example 5
This embodiment provides a detection system of hotel static information, as shown in fig. 4, the detection system includes: a first feedback information acquisition module 401, a second feedback information acquisition module 402, a hotel static information acquisition module 403, and a static information judgment module 404.
The first feedback information acquisition module 401 is configured to acquire first feedback information of a user on a hotel, where the first feedback information includes comment information of the user on the hotel;
the second feedback information obtaining module 402 is configured to input the first feedback information into a static information prediction model to obtain second feedback information including hotel static information, where the static information prediction model is obtained after a machine learning model is trained based on historical feedback information, and the historical feedback information includes historical comment information of a user on a hotel;
the hotel static information obtaining module 403 is configured to obtain hotel static information provided by the hotel;
the static information determining module 404 is configured to determine whether content of hotel static information in the second feedback information is consistent with content of corresponding hotel static information provided by the hotel, and if not, check the inconsistent static information.
The first feedback information comprises comment information of the user on the hotel.
The static information prediction model is obtained after a machine learning model is trained on the basis of historical feedback information, the historical feedback information comprises historical comment information of a user on a hotel, a deep learning text classification method in the machine learning model needs to collect text corpora in advance to construct Word vectors of texts, and then the deep learning classification algorithm Word vectors are used on the basis of existing Word vectors, and the Word vectors are constructed by methods such as Word2Vec and GloVe.
According to the embodiment, the comment condition of the user on the hotel is fully utilized through the first feedback information acquisition module, the problem possibly existing in the hotel static information provided by the hotel is timely excavated through the second feedback information acquisition module, the hotel static information fed back by the user is compared with the hotel static information provided by the hotel through the static information judgment module, and therefore whether the hotel static information provided by the hotel is correct or not is judged, if not, error correction is timely carried out, and the accuracy of the hotel static information is improved.
Example 6
In this embodiment, the static information determining module 404 is further configured to determine whether the number of second feedback information inconsistent with the corresponding hotel static information provided by the hotel exceeds an error correction information threshold, if so, determine whether the hotel static information provided by the hotel is consistent with the real static information of the hotel, and if not, modify the hotel static information provided by the hotel into the real static information of the hotel.
In the embodiment, the influence of malicious comment on the detection system can be avoided by setting the error correction threshold in the static information judgment module, so that the accuracy of detecting the hotel static information provided by the hotel is improved.
In this embodiment, the static information judgment module can in time update inconsistent static information that the hotel provided to promote the accuracy of the static information of hotel, promote user's experience of living in, reduce the amount of complaint of user to the hotel, promote OTA's brand image.
Example 7
This example is a further modification of example 5 or example 6, and in this example,
the first feedback information also comprises complaint information of the user to the hotel, dialogue information of the user and customer service and other related information which can be obtained and fed back to the hotel by the user.
The embodiment further includes a model obtaining module, configured to obtain the static information prediction model in the second feedback information obtaining module 402.
As shown in fig. 5, the model obtaining module specifically includes: a history feedback information obtaining unit 501, a history information selecting unit 502, a preprocessing unit 503, a selected information judging unit 504, a dividing unit 505, a training unit 506, a testing unit 507, and an index judging unit 508.
The history feedback information obtaining unit 501 is configured to obtain history feedback information of the user on the hotel, and call the history information selecting unit 502.
The history information selecting unit 503 is configured to select history feedback information of which the number is a preset number, and call the preprocessing unit 503.
The preprocessing unit 503 is configured to preprocess the historical feedback information, and call the selected information determining unit 504, where the preprocessing includes at least one of removing sensitive words, removing special characters, converting full angles to half angles, converting traditional characters to simplified characters, and converting upper case to lower case.
The selected information determining unit 504 is configured to determine whether the selected historical feedback information includes the hotel static information, if so, mark the historical feedback information as a first machine code, otherwise, mark the historical feedback information as a second machine code, and call the dividing unit 505.
The dividing unit 505 is configured to divide the marked historical feedback information into a training set and a test set, and invoke the training unit 506.
The training unit 506 is configured to train the historical feedback information in the training set as input and the first machine code and the second machine code corresponding to the labels as output in the machine learning model, and invoke the testing unit 507.
The testing unit 507 is configured to input the historical feedback information in the test set into the trained machine learning model to obtain a testing machine code corresponding to each piece of the historical feedback information, and call the index determining unit 508.
The index determining unit 508 is configured to determine whether an evaluation index AUC of the test machine code reaches an evaluation threshold, and if so, the trained machine learning model is the static information prediction model.
The machine learning model can be a deep learning model which can be acquired by TextCNN, LSTM, BI-LSTM and the like.
The evaluation index AUC is an area under an ROC curve, an abscissa of the area under the ROC curve is a ratio of the number of second machine codes to the number of test machine codes inconsistent with the number of corresponding marked machine codes, and an ordinate of the area under the ROC curve is a ratio of the number of first machine codes to the number of first machine codes consistent with the number of corresponding marked machine codes.
For better understanding of the present embodiment, the present embodiment will be further explained by taking a specific example as follows:
firstly, acquiring comment content of a user on a hotel in the OTA industry, conversation text content of the user and hotel customer service, and common open-source text corpora such as some common newspapers and web news, constructing word vectors for a deep learning text classification method, and constructing the dimension of each obtained word vector to be 100-dimensional or 200-dimensional. And then, extracting historical feedback information of the user on the hotel, and then carrying out pretreatment such as special character removal, full angle turning to half angle, traditional turning to simplified form, upper writing to lower writing and the like on the historical feedback information. And then, selecting a part of the processed historical feedback information for manual marking, and judging whether the processed historical feedback information has the content of static information, if so, marking the processed historical feedback information as 1, otherwise, marking the processed historical feedback information as 0, and if the processed historical feedback information is: when the hotel environment is poor, the piece of content does not include static information, and is marked as 0, and when the feedback information is: the hotel has a parking lot, and since the parking lot is static information, the label of the piece of feedback information is 0. Then, the marked content is randomly divided into a training set A and a testing set B according to a certain proportion (such as 7: 3). Deep learning model training is carried out on historical feedback information in the training set A, the model is verified by using data in the test set B, when the evaluation index AUC of the static information problems in the verified test set reaches a threshold value, the training is stopped to obtain the model, otherwise, the model is continuously trained, wherein the higher the AUC index is, the better the model effect is represented. After the model is trained, the first feedback information is input into the trained model, and in this embodiment, the first feedback information includes 5 pieces of content: the hotel has no delivery service, the hotel environment is elegant, the swimming pool of the hotel is not wrong, the attitude of the customer service staff is poor, the hotel does not provide the swimming pool to obtain the content with the static information problem, and the obtained content is the content with the static information, in the embodiment, the following three contents are obtained: the hotel has no pick-up service, the swimming pool of the hotel is good, and the hotel does not provide the swimming pool. Next, whether feedback information inconsistent with the content of the static information provided by the hotel reaches an error correction information threshold value is determined, in this embodiment, the set error correction threshold value is 5, and only 2 pieces of information of the hotel with no delivery service are provided as the feedback information, but 12 pieces of information of the hotel with no swimming pool are provided as the feedback information, the hotel with delivery service is temporarily defaulted, and for the feedback of the hotel with no swimming pool, the content of the static information provided by the hotel is checked, and the static information provided by the hotel is checked as follows: the hotel has a swimming pool, the feedback information provides no swimming pool for the hotel, the information is inconsistent, then the next step is carried out, whether the static information of the hotel provided by the pitted real hotel is consistent with the real static information of the hotel, and if the real situation of the hotel is that no swimming pool exists, the static information of the hotel can be corrected.
In the embodiment, the historical feedback information is input into the machine model through the model acquisition module to be trained to obtain the static information prediction model, so that whether the static information exists in the feedback information can be screened through a model method, the workload of manually distinguishing whether each piece of the large amount of feedback information has the static information can be greatly reduced, the labor force is greatly saved, and the detection efficiency is improved.
In this embodiment, the AUC indexes set by the index determination unit can avoid the situation that the model training is not accurate and the predicted static information is not accurate, and can further train a more accurate static information prediction model.
In this embodiment, when the determining unit determines that the AUC index does not reach the threshold, the training unit may update the model in real time by retraining the model, so as to obtain a more accurate static information prediction model.
In this embodiment, the static information judgment module can in time update inconsistent static information that the hotel provided to promote the accuracy of the static information of hotel, promote user's experience of living in, reduce the amount of complaint of user to the hotel, promote OTA's brand image.
In the embodiment, the feedback information of the hotel is fully utilized by utilizing a classification method of deep learning, the feedback information is predicted through the model, the content reflecting the problem of the static information is obtained, the static information item with the problem is extracted to be matched and compared with the static information item provided by the hotel, inconsistent content is mined to be manually verified, the wrong static information content in the hotel is corrected in time, the comment prediction is carried out by using the model method, and the manual judgment of comment texts can be greatly reduced. Meanwhile, through manual verification of the hotel with problems, the cost for manually verifying all hotels once is greatly reduced, the accuracy of static information of the hotel is improved, the check-in experience of a user is improved, the complaint amount of the user to the hotel is reduced, and the brand image of the OTA is improved.
Example 8
This embodiment is basically the same as embodiment 7 except that the historical feedback information in embodiment 7 is replaced with historical point score information having a point score lower than a point score threshold.
The point score threshold value can be a median of a hotel point score range, for example, if the value of the hotel point score is 1-5 points, the threshold value is 3 points.
Because the historical comment information with the point score lower than the point score threshold generally reflects that the content of the static information item of the hotel is in a problem, in the embodiment, the training unit trains the historical comment information with the point score lower than the point score threshold, and the static information provided by the hotel and inconsistent with the real situation can be acquired more pertinently according to the negative feedback of the user, so that the relevant hotel can check the static information conveniently, and the accuracy of the static information provided by the hotel is improved.
Example 9
The present embodiment provides an electronic device, which may be represented in the form of a computing device (for example, may be a server device), including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the detection methods of embodiments 1 to 4.
Fig. 6 shows a schematic diagram of a hardware structure of the present embodiment, and as shown in fig. 6, the electronic device 9 specifically includes:
at least one processor 91, at least one memory 92, and a bus 93 for connecting the various system components (including the processor 91 and the memory 92), wherein:
the bus 93 includes a data bus, an address bus, and a control bus.
The processor 91 executes various functional applications and data processing, such as the detection methods provided in embodiments 1-4 of the present invention, by running a computer program stored in the memory 92.
The electronic device 9 may further communicate with one or more external devices 94 (e.g., a keyboard, a pointing device, etc.). Such communication may be through an input/output (I/O) interface 95. Also, the electronic device 9 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet) via the network adapter 96. The network adapter 96 communicates with the other modules of the electronic device 9 via the bus 93. It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 9, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.
It should be noted that although in the above detailed description several units/modules or sub-units/modules of the electronic device are mentioned, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the units/modules described above may be embodied in one unit/module, according to embodiments of the application. Conversely, the features and functions of one unit/module described above may be further divided into embodiments by a plurality of units/modules.
Example 10
The present embodiment provides a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements the steps of the detection method provided in embodiments 1-4.
More specific examples, among others, that the readable storage medium may employ may include, but are not limited to: a portable disk, a hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device, or any suitable combination of the foregoing.
In a possible implementation, the invention can also be implemented in the form of a program product comprising program code for causing a terminal device to carry out the steps of implementing the detection method in embodiments 1 to 4, when said program product is run on said terminal device.
Where program code for carrying out the invention is written in any combination of one or more programming languages, the program code may be executed entirely on the user device, partly on the user device, as a stand-alone software package, partly on the user device and partly on a remote device or entirely on the remote device.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.
Claims (12)
1. A detection method for hotel static information is characterized by comprising the following steps:
acquiring first feedback information of a user on a hotel, wherein the first feedback information comprises comment information of the user on the hotel;
inputting the first feedback information into a static information prediction model to obtain second feedback information comprising hotel static information, wherein the static information prediction model is obtained after training a machine learning model based on historical feedback information, and the historical feedback information comprises historical comment information of a user on the hotel;
acquiring hotel static information provided by the hotel;
and judging whether the content of the hotel static information in the second feedback information is consistent with the content of the corresponding hotel static information provided by the hotel, and if not, checking the inconsistent static information.
2. The detection method of claim 1, wherein the step of verifying the static information that is inconsistent comprises:
judging whether the quantity of second feedback information inconsistent with the corresponding hotel static information provided by the hotel exceeds an error correction information threshold value, and if so, verifying the corresponding hotel static information provided by the hotel;
and/or the presence of a gas in the gas,
the first feedback information also comprises complaint information of the user to the hotel and dialogue information of the user and customer service.
3. The detection method of claim 1, further comprising:
obtaining the static information prediction model by:
acquiring historical feedback information of a user on a hotel;
selecting historical feedback information with the quantity of a preset numerical value;
judging whether the selected historical feedback information comprises the hotel static information, if so, marking the historical feedback information as a first machine code, and if not, marking the historical feedback information as a second machine code;
dividing the marked historical feedback information into a training set and a test set;
taking the historical feedback information in the training set as input and taking the first machine code and the second machine code which are correspondingly marked as output, and training the machine codes in a machine learning model;
inputting the historical feedback information in the test set into a trained machine learning model to obtain a test machine code corresponding to each piece of historical feedback information;
and judging whether the evaluation index AUC of the test machine code reaches an evaluation threshold value, if so, the trained machine learning model is the static information prediction model.
4. The detection method according to claim 3,
if the evaluation index AUC is judged not to reach the evaluation threshold value, adding the marked new historical feedback information into the training set for training again;
and/or the presence of a gas in the gas,
the step of selecting the historical feedback information with the quantity of the preset numerical value further comprises the following steps: preprocessing the historical feedback information, wherein the preprocessing comprises at least one of sensitive word removal, special character removal, full angle conversion of half angle, traditional Chinese character conversion to simple Chinese character conversion and upper case conversion to lower case conversion;
and/or the presence of a gas in the gas,
the historical feedback information is historical comment information with the comment score lower than a comment score threshold value.
5. The detection method according to claim 1,
the step of verifying the static information as inconsistent comprises: judging whether the hotel static information provided by the hotel is consistent with the real static information of the hotel, if not, modifying the hotel static information provided by the hotel into the real static information of the hotel;
and/or the presence of a gas in the gas,
the machine training model is a deep learning model.
6. A detection system for hotel static information, the detection system comprising:
the system comprises a first feedback information acquisition module, a second feedback information acquisition module, a hotel static information acquisition module and a static information judgment module;
the first feedback information acquisition module is used for acquiring first feedback information of a user on the hotel, and the first feedback information comprises comment information of the user on the hotel;
the second feedback information acquisition module is used for inputting the first feedback information into a static information prediction model to acquire second feedback information comprising hotel static information, the static information prediction model is obtained after a machine learning model is trained on the basis of historical feedback information, and the historical feedback information comprises historical comment information of a user on the hotel;
the hotel static information acquisition module is used for acquiring hotel static information provided by the hotel;
the static information judgment module is used for judging whether the content of the hotel static information in the second feedback information is consistent with the content of the corresponding hotel static information provided by the hotel, and if not, the inconsistent static information is verified.
7. The detection system of claim 6, wherein the static information determination module is further configured to determine whether a quantity of second feedback information inconsistent with the corresponding hotel static information provided by the hotel exceeds an error correction information threshold, and if so, verify the corresponding hotel static information provided by the hotel;
and/or the presence of a gas in the gas,
the first feedback information also comprises complaint information of the user to the hotel and dialogue information of the user and customer service.
8. The detection system of claim 6, further comprising:
the model obtaining module is used for obtaining the static information prediction model;
the model acquisition module comprises: the device comprises a historical feedback information acquisition unit, a historical information selection unit, a selection information judgment unit, a division unit, a training unit, a test unit and an index judgment unit;
the historical feedback information acquisition unit is used for acquiring historical feedback information of a user on the hotel and calling the historical information selection unit;
the history information selection unit is used for selecting history feedback information with the number of preset values and calling the selected information judgment unit;
the selected information judging unit is used for judging whether the selected historical feedback information comprises the hotel static information, if so, the historical feedback information is marked as a first machine code, if not, the historical feedback information is marked as a second machine code, and the dividing unit is called;
the dividing unit is used for dividing the marked historical feedback information into a training set and a test set and calling the training unit;
the training unit is used for taking the historical feedback information in the training set as input and taking the first machine code and the second machine code which are correspondingly marked as output, training the historical feedback information in a machine learning model, and calling the testing unit;
the test unit is used for inputting the historical feedback information in the test set into a trained machine learning model to obtain a test machine code corresponding to each piece of historical feedback information, and calling the index judgment unit;
the index judgment unit is used for judging whether the evaluation index AUC of the test machine code reaches an evaluation threshold value, and if the evaluation index AUC reaches the evaluation threshold value, the trained machine learning model is the static information prediction model.
9. The detection system of claim 8,
if the index judgment unit judges that the evaluation index AUC does not reach an evaluation threshold value, the training unit is called and is also used for adding the marked new historical feedback information into the training set for training again;
and/or the presence of a gas in the gas,
the model acquisition module further comprises: the preprocessing unit is used for preprocessing the historical feedback information and comprises at least one of sensitive word removal, special character removal, full angle conversion of half angle, traditional Chinese character conversion to simplified Chinese character conversion and upper case conversion to lower case conversion;
and/or the presence of a gas in the gas,
the historical feedback information is historical comment information with the comment score lower than a comment score threshold value.
10. The detection system of claim 6, wherein the static information determination module is further configured to determine whether hotel static information provided by the hotel is consistent with the real static information of the hotel, and if not, modify the hotel static information provided by the hotel into the real static information of the hotel;
and/or the presence of a gas in the gas,
the machine training model is a deep learning model.
11. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the detection method of any one of claims 1 to 5 when executing the computer program.
12. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the detection method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911142094.5A CN110930022A (en) | 2019-11-20 | 2019-11-20 | Hotel static information detection method and system, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911142094.5A CN110930022A (en) | 2019-11-20 | 2019-11-20 | Hotel static information detection method and system, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110930022A true CN110930022A (en) | 2020-03-27 |
Family
ID=69851348
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911142094.5A Pending CN110930022A (en) | 2019-11-20 | 2019-11-20 | Hotel static information detection method and system, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110930022A (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324628A (en) * | 2012-03-21 | 2013-09-25 | 腾讯科技（深圳）有限公司 | Industry classification method and system for text publishing |
| CN106295831A (en) * | 2016-08-29 | 2017-01-04 | 徐月明 | Guest room reservation method and system and hotel information management platform |
| CN106875134A (en) * | 2017-03-30 | 2017-06-20 | 携程计算机技术（上海）有限公司 | The Min Su inns method for digging of OTA |
| CN108665283A (en) * | 2018-04-28 | 2018-10-16 | 携程计算机技术（上海）有限公司 | The recognition methods of hotel's house type price exception of OTA platforms and system |
| CN108763199A (en) * | 2018-05-14 | 2018-11-06 | 浙江口碑网络技术有限公司 | The investigation method and device of text feedback information |
| US20180357570A1 (en) * | 2017-06-13 | 2018-12-13 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for building a conversation understanding system based on artificial intelligence, device and computer-readable storage medium |
| CN109858683A (en) * | 2019-01-07 | 2019-06-07 | 北京三快在线科技有限公司 | Determine method, apparatus, electronic equipment and the storage medium of the business status in shop |
| CN109918501A (en) * | 2019-01-18 | 2019-06-21 | 平安科技（深圳）有限公司 | Method, apparatus, equipment and the storage medium of news article classification |
| WO2019174395A1 (en) * | 2018-03-13 | 2019-09-19 | 阿里巴巴集团控股有限公司 | Method and apparatus for information recommendation, and device |
| CN110347840A (en) * | 2019-07-18 | 2019-10-18 | 携程计算机技术（上海）有限公司 | Complain prediction technique, system, equipment and the storage medium of text categories |
-
2019
- 2019-11-20 CN CN201911142094.5A patent/CN110930022A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324628A (en) * | 2012-03-21 | 2013-09-25 | 腾讯科技（深圳）有限公司 | Industry classification method and system for text publishing |
| CN106295831A (en) * | 2016-08-29 | 2017-01-04 | 徐月明 | Guest room reservation method and system and hotel information management platform |
| CN106875134A (en) * | 2017-03-30 | 2017-06-20 | 携程计算机技术（上海）有限公司 | The Min Su inns method for digging of OTA |
| US20180357570A1 (en) * | 2017-06-13 | 2018-12-13 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for building a conversation understanding system based on artificial intelligence, device and computer-readable storage medium |
| WO2019174395A1 (en) * | 2018-03-13 | 2019-09-19 | 阿里巴巴集团控股有限公司 | Method and apparatus for information recommendation, and device |
| CN108665283A (en) * | 2018-04-28 | 2018-10-16 | 携程计算机技术（上海）有限公司 | The recognition methods of hotel's house type price exception of OTA platforms and system |
| CN108763199A (en) * | 2018-05-14 | 2018-11-06 | 浙江口碑网络技术有限公司 | The investigation method and device of text feedback information |
| CN109858683A (en) * | 2019-01-07 | 2019-06-07 | 北京三快在线科技有限公司 | Determine method, apparatus, electronic equipment and the storage medium of the business status in shop |
| CN109918501A (en) * | 2019-01-18 | 2019-06-21 | 平安科技（深圳）有限公司 | Method, apparatus, equipment and the storage medium of news article classification |
| CN110347840A (en) * | 2019-07-18 | 2019-10-18 | 携程计算机技术（上海）有限公司 | Complain prediction technique, system, equipment and the storage medium of text categories |
Non-Patent Citations (1)
| Title |
|---|
| 陈东明 等: "《Python编程及其在网络科学中的应用》", 30 November 2018, 东北大学出版社 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11449379B2 (en) | Root cause and predictive analyses for technical issues of a computing environment | |
| CN106202453B (en) | Multimedia resource recommendation method and device | |
| US20190333118A1 (en) | Cognitive product and service rating generation via passive collection of user feedback | |
| CN110674255B (en) | Text content auditing method and device | |
| CN105989268A (en) | Safety access method and system for human-computer identification | |
| CN111181757B (en) | Information security risk prediction method and device, computing equipment and storage medium | |
| CN110647523B (en) | Data quality analysis method and device, storage medium and electronic equipment | |
| CN111144118B (en) | Method, system, equipment and medium for identifying named entities in spoken text | |
| CN112491649A (en) | Interface joint debugging test method and device, electronic equipment and storage medium | |
| CN110555488A (en) | Image sequence auditing method and system, electronic equipment and storage medium | |
| US10489728B1 (en) | Generating and publishing a problem ticket | |
| EP4285272A1 (en) | Root cause pattern recognition based model training | |
| CN113190746A (en) | Recommendation model evaluation method and device and electronic equipment | |
| CN113705164A (en) | Text processing method and device, computer equipment and readable storage medium | |
| CN114757186B (en) | User intention analysis method and device, computer storage medium and electronic equipment | |
| US20230221847A1 (en) | Cognitive detection of user interface errors | |
| CN115292187A (en) | Method and device for automatically testing code-free page, electronic equipment and medium | |
| CN110930022A (en) | Hotel static information detection method and system, electronic equipment and storage medium | |
| CN118339550A (en) | Geometric problem solving method, device, equipment and storage medium | |
| CN114443493A (en) | Test case generation method and device, electronic equipment and storage medium | |
| CN110083807B (en) | Contract modification influence automatic prediction method, device, medium and electronic equipment | |
| CN113627197A (en) | Text intention recognition method, device, equipment and storage medium | |
| CN113052509A (en) | Model evaluation method, model evaluation apparatus, electronic device, and storage medium | |
| KR20160088737A (en) | System and Method for Bug Fixing Developers Recommendation and Bug Severity Prediction based on Topic Model and Multi-Feature | |
| US20200311776A1 (en) | Transforming content management in product marketing |
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: 20200327 |