A method for constructing a chemotherapy-related oral mucositis nursing knowledge base

Abstract

The present application relates to the technical field of nursing knowledge base construction, in particular to a construction method of a chemotherapy-related oral mucositis nursing knowledge base. The present application first acquires a knowledge base and access data; further acquires an initial index weight according to the recent access trend and the total access times of each keyword, in combination with the distribution of keywords in clinical nursing data; further acquires nursing pertinence according to the difference between the clinical nursing data involved by each keyword, in combination with the recovery effect; and finally acquires the final index weight of each keyword by fusing the initial index weight and the nursing pertinence. The present scheme dynamically optimizes the keyword index weight by fusing the universality, timeliness and nursing pertinence of keywords, takes into account the extensive demand and individual efficacy, avoids the virtual high of hot and inefficient words, and improves the retrieval accuracy and the application value of the database.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of nursing knowledge base construction, and in particular to a construction method of a chemotherapy-related oral mucositis nursing knowledge base. BACKGROUND

[0002] Oral mucositis caused by chemotherapy is called chemotherapy-induced oral mucositis (CTOM), and appropriate nursing knowledge is used for nursing of tumor patients during the CTOM period, which can effectively promote the healing of CTOM of the patients and avoid further deterioration and affect the course of tumor treatment.

[0003] In the prior art, the clinical nursing data of patients needing chemotherapy who suffer from CTOM is used to construct a chemotherapy-related oral mucositis nursing knowledge base, so that medical staff and patients themselves can better understand the relevant nursing knowledge and effectively promote the healing of CTOM. However, due to the large amount of knowledge base data, when indexing the nursing knowledge, all the nursing knowledge data matching the index are listed, and the user cannot quickly find the nursing knowledge he wants to index, and the indexing efficiency is low. SUMMARY

[0004] In order to solve the technical problem of low efficiency of the indexing method of the existing chemotherapy-related oral mucositis nursing knowledge base, the purpose of the present application is to provide a construction method of a chemotherapy-related oral mucositis nursing knowledge base, and the technical solution adopted is as follows:

[0005] Obtain a knowledge base constructed by historical patient clinical nursing data and access data, and extract keywords and recovery effects in the clinical nursing data;

[0006] According to the recent access trend and the total access times of each keyword in the access data, and in combination with the distribution of the keyword in the clinical nursing data, an initial index weight of each keyword is obtained;

[0007] According to the difference between the clinical nursing data related to each keyword, in combination with the recovery effect, the nursing specificity of each keyword is obtained; and the initial index weight and the nursing specificity are fused to obtain the final index weight of each keyword.

[0008] Further, the method for obtaining the initial index weight comprises:

[0009] According to the distribution of each keyword in the clinical nursing data, a distribution weight is obtained;

[0010] The access data further includes the single-day access times of each keyword; and according to the change of the single-day access times of each keyword, a recent access intensity is obtained;

[0011] Fusing the distribution weight, the recent access intensity and the total access times, an initial index weight of each keyword is obtained.

[0012] Further, the method for obtaining the distribution weight comprises:

[0013] Taking the number of occurrences of each keyword in the knowledge base as the respective distribution weight.

[0014] Further, the method for obtaining the recent access intensity comprises:

[0015] For each keyword, a slope of a fitting straight line of the single-day access times in a preset historical neighborhood of the current day is obtained, and a highlight feature of the maximum single-day access time in the preset historical neighborhood of the current day is combined to obtain the recent access intensity.

[0016] Further, the method for obtaining the nursing targeting comprises:

[0017] Patient data in the clinical nursing data is vectorized, and a data vector and the recovery effect form a binary tuple.

[0018] For each keyword, in all the binary tuples of the involved clinical nursing data: the binary tuple with the maximum recovery effect is selected as a target binary tuple, a difference between the recovery effect of the target binary tuple and other binary tuples is obtained, and a difference between the data vectors is combined to obtain a nursing targeting factor.

[0019] The nursing targeting factor and the recovery effect of the target binary tuple are fused to obtain a nursing targeting of the corresponding keyword.

[0020] Further, the method for obtaining the data vector comprises:

[0021] Patient data is vectorized by using one-hot encoding to obtain a data vector.

[0022] Further, the method for obtaining the final index weight comprises:

[0023] A product of the initial index weight of each keyword and the nursing targeting is taken as the final index weight.

[0024] Further, after obtaining the final index weight of each keyword, the method further comprises:

[0025] An index keyword input by a current user is obtained, and a knowledge base system obtains a series of keyword forms matched with the index keyword input by the user by using a string matching algorithm; the keyword forms are arranged in descending order of the final index weight, and the arranged keyword forms are output.

[0026] Further, the length of the preset history neighborhood is 7 days.

[0027] Further, the keyword acquisition method comprises:

[0028] The nursing knowledge topic field in the preset nursing knowledge sheet is set as a key field, and the key field involved in the clinical nursing data is identified and taken as a keyword.

[0029] The present application has the following beneficial effects:

[0030] The present application firstly acquires a knowledge base and access data, providing an analysis basis for the subsequent; further, according to the recent access trend and the total access times of each keyword, combining the distribution of the keyword in the clinical nursing data, while taking into account the use of the historical nursing data, the universality of the keyword in the actual use, and the timeliness of the keyword in the recent clinical nursing demand, an initial index weight is acquired, improving the accuracy of the initial index weight; further, in order to avoid invalid access of part of the keyword, according to the difference between the clinical nursing data involved by each keyword, combining the recovery effect, nursing pertinence is acquired, providing a basis for correcting the initial index weight from the perspective of the pertinence of the nursing knowledge; finally, the initial index weight and the nursing pertinence are fused to acquire the final index weight of each keyword. The present application dynamically optimizes the keyword index weight by fusing the universality, timeliness and nursing pertinence of the keyword, taking into account the extensive demand and individual curative effect, avoiding the virtual high of the popular inefficient word, and improving the retrieval accuracy and the application value of the database. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0032] Figure 1 The flow chart of the construction method of the chemotherapy related oral mucositis nursing knowledge base provided by an embodiment of the present application;

[0033] Figure 2 The flow chart of the initial index weight acquisition method provided by an embodiment of the present application;

[0034] Figure 3 The flow chart of the nursing pertinence acquisition method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to further clarify the technical means and effects taken by the present application to achieve the predetermined inventive objectives, the following describes in detail the specific implementation, structure, features and effects of the method for constructing a chemotherapy-related oral mucositis nursing knowledge base according to the present application, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0037] The following describes in detail the specific scheme of the method for constructing a chemotherapy-related oral mucositis nursing knowledge base according to the present application.

[0038] Please refer to Figure 1 which shows a flowchart of the method for constructing a chemotherapy-related oral mucositis nursing knowledge base according to one embodiment of the present application, which specifically includes:

[0039] Step S1: Obtain the knowledge base constructed by historical patient clinical nursing data and access data, and extract keywords and recovery effects from the clinical nursing data.

[0040] In one embodiment of the present application, the clinical nursing data at least includes the patient's nursing drug, the drug type used for chemotherapy treatment, the oral mucositis onset and recovery time (relative to the time after the first treatment), the recovery effect of nursing condition on CTMO, the patient's disease history, and the drug allergy history. The clinical nursing data is arranged in a corresponding order, for example, [rehabilitation new liquid; paclitaxel, cyclophosphamide; 6d; 17d; 0.65; breast cancer; no allergy history], which means that the patient's nursing drug is rehabilitation new liquid; the drug type used for chemotherapy treatment is paclitaxel and cyclophosphamide; the oral mucositis onset is on the 6th day after the first treatment, and the recovery is on the 17th day; the recovery effect is 0.65, there is a history of breast cancer, and there is no drug allergy history.

[0041] After the clinical nursing data is converted into a structured object (such as JSON, dictionary), it is stored in the database to construct a chemotherapy-related oral mucositis nursing knowledge base, which is referred to as a knowledge base, and access data of the knowledge base is obtained.

[0042] Further, in order to facilitate the user to index the database, the keywords in the clinical nursing data are extracted.

[0043] Preferably, in one embodiment of the present application, the nursing knowledge topic field in the preset nursing knowledge single is set as a key field, and the key field involved in the clinical nursing data is identified as a keyword.

[0044] As an example, the preset nursing knowledge sheet is the Expert Consensus on Prevention and Treatment of Chemotherapy-induced Oral Mucositis (T / CHSA 074-2024), which contains various recommended detailed nursing knowledge such as oral management, rehabilitation new liquid, etc. The nursing knowledge topic field is set as a key field for indexing matching, identifying the key field involved in the clinical nursing data and serving as a keyword. The keyword is matched with the detailed nursing knowledge corresponding thereto and the clinical nursing data using the nursing knowledge.

[0045] Then, the index weight of the keyword is established, and the index module of the knowledge base is constructed.

[0046] It should be noted that the recovery effect is a specific data value, and the higher the data value, the better the recovery of the patient. The data of the historical patients used are desensitized and do not involve sensitive information such as the name and personal basic information of the patient. The construction method of the knowledge base is a prior art and will not be described in detail.

[0047] In other embodiments of the present application, the implementer can adjust the structure and content of the clinical nursing data; and the MASCC / ISOO clinical practice guidelines for the management of mucositis secondary to cancer therapy can also be set as a preset nursing knowledge sheet.

[0048] Step S2: According to the recent access trend and the total access times of each keyword in the access data, and in combination with the distribution of the keyword in the clinical nursing data, the initial index weight of each keyword is obtained.

[0049] It is considered that the distribution of the keyword in the clinical nursing data reflects the selection in the historical nursing record; after the knowledge base is put into use, the user can access the nursing knowledge mapped by different keywords through keyword indexing; the access data records the total access times, reflecting the universality of the selection of the keyword; and the knowledge also has timeliness, and the patient's symptoms also change, so the recent access trend of the keyword can reflect the applicability of the nursing knowledge mapped thereby in a recent period of time.

[0050] Therefore, according to the recent access trend and the total access times of each keyword in the access data, and in combination with the distribution of the keyword in the clinical nursing data, the initial index weight of each keyword is obtained, while the use of the historical nursing data, the universality of the keyword in actual use, and the timeliness of the keyword in the recent clinical nursing demand are also taken into account, so that the index result is more in line with the retrieval intention of the current user, and the precision of keyword matching and the index efficiency of the knowledge base are significantly improved.

[0051] Preferably, in one embodiment of the present application, referring to Figure 2 , which shows a flow chart of an initial index weight acquisition method provided by one embodiment of the present application, and specifically comprises:

[0052] Step S201: According to the distribution of each keyword in the clinical nursing data, the distribution weight is acquired.

[0053] It is considered that the more times a keyword appears in the clinical nursing data in the knowledge base, the more times the corresponding nursing operation is selected in the nursing scheme for chemotherapy-related oral mucositis, and the stronger the universality is. Therefore, the number of times each keyword appears in the knowledge base is taken as the respective distribution weight, so as to show the distribution characteristics of the keyword in the clinical nursing data.

[0054] In another embodiment of the present application, the number of times each keyword appears in the database, the proportion of the total number of times all keywords appear, is taken as the distribution weight; it is to be noted that the final sorting of the index weight is based on the relative size between keywords, so the calculation method of the distribution weight of all keywords can be unified.

[0055] Step S202: The access data further includes the single-day access number of each keyword; according to the change of the single-day access number of each keyword, the recent access intensity is acquired.

[0056] In one embodiment of the present application, the daily access amount is recorded in days, the access data further includes the single-day access number of each keyword, the recent access trend of the keyword is shown by means of the change of the single-day access number, the recent access intensity of the keyword is acquired, and the trend intensity is represented.

[0057] It is considered that the stronger the increasing trend of the single-day access number in the current historical neighborhood is, the higher the recent access intensity is, and the trend intensity is rising; at the same time, the more prominent the maximum single-day access number in the historical neighborhood is, the higher the recent access intensity is, and there is a large amount of access to the corresponding keyword in the recent period. Based on this, for each keyword, according to the slope of the fitting straight line of the single-day access number in the preset historical neighborhood of the current day, in combination with the prominent feature of the maximum single-day access number in the preset historical neighborhood of the current day, the recent access intensity is acquired.

[0058] As an example, the time length of the preset historical neighborhood is 7 days, for each keyword, the single-day access number of the adjacent 7 days of the current day is taken, a fitting straight line of the single-day access number is obtained based on the least square method, the slope is obtained, the maximum single-day access number in the preset historical neighborhood of the current day is taken as the numerator, the maximum single-day access number in all historical days before the preset historical neighborhood of the current day is taken as the denominator, the mapping value of the slope after being mapped by an exponential function exp(x) with the natural constant e as the base is taken as the product of the fraction ratio, and the product is taken as the recent access intensity of the corresponding keyword.

[0059] In order to prevent the influence of the slope being zero or negative on the logical relationship, the slope is mapped by an exponential function exp(x) with the natural constant e as the base, and x is the independent variable; the maximum single-day access number in the preset historical neighborhood of the current day is taken as the reference, and the maximum single-day access number in the preset historical neighborhood of the current day is highlighted by the fraction ratio, the greater the fraction ratio, the closer the recent access of the keyword to the historical peak access, and the stronger the highlighting feature.

[0060] The fitting straight line of the least square method and the acquisition method of the slope of the straight line are both prior art, when a certain keyword is a new word, the historical days are less than 8 days but at least 2 days, that is, the highlighting feature of the maximum single-day access number in the preset historical neighborhood of the current day cannot be analyzed, the slope of the fitting straight line of the single-day access number of the existing historical days is directly taken as the mapping value after being mapped by the exponential function exp(x) with the natural constant e as the base, and the mapping value is taken as the recent access intensity of the corresponding keyword.

[0061] When the historical days are less than 2 days, the average value of the latest recent access intensity of other keywords is taken as the recent access intensity of the current keyword.

[0062] Step S203: Fusion distribution weight, recent access intensity and total access number to obtain the initial index weight of each keyword.

[0063] Finally, it is considered that the more the total access number of the keyword history, the greater the possibility of being accessed again, and the greater the initial index weight; the greater the distribution weight, the stronger the universality, and the greater the initial index weight; the greater the recent access intensity, the greater the recent access, the higher the possibility of being accessed recently, and the greater the initial index weight, and finally the three are fused to obtain the initial index weight of each keyword.

[0064] As an example, the product of the distribution weight, the recent access intensity and the total access number of each keyword in the current day is taken as the initial index weight of each keyword.

[0065] Wherein, the total access times and the distribution weight are data characteristics of all historical days before the current day; as another example, the distribution weight, the recent access intensity and the total access times are linearly normalized in the respective corresponding data dimensions respectively, and the three normalized results are fused by weighted summation, and the weights can be 1 / 3, 1 / 3 and 1 / 3.

[0066] It should be noted that the acquisition manner of the initial index weight of each keyword is consistent, and only one example is described here, and the repeated description is not repeated.

[0067] Step S3: According to the difference between the clinical nursing data involved by each keyword, the recovery effect is combined to obtain the nursing pertinence of each keyword; the initial index weight and the nursing pertinence are fused to obtain the final index weight of each keyword.

[0068] However, most users will prefer to access the keywords arranged in front (i.e. the initial index weight is large) after indexing, and the total number of views is high, and the nursing knowledge corresponding to these keywords is usually a relatively basic nursing method, which is highly universal but often has poor effect.

[0069] Some keywords with a large number of access times in a short time have a large number of invalid accesses due to the snowball effect, and their historical access times gradually increase, resulting in that the index weight value obtained by only referring to the historical access situation is also gradually high, which in turn makes the user invalid access the keyword after indexing, and the access times become large, forming a vicious cycle. Therefore, it is necessary to further refer to the effectiveness of the keyword to further modify the initial index weight.

[0070] In the clinical nursing data, different nursing knowledge promotes different CTOM recovery effects of different individual patients, and different individuals often need different targeted nursing knowledge. The stronger the pertinence to the patient is, the more likely it is to be a nursing knowledge method that some users do not know, and the more likely it is to solve the difficulties of users and become the expected index target of users, and the effective access probability of users needs to be increased.

[0071] Therefore, according to the difference between the clinical nursing data involved by each keyword, the pertinence characteristics of the nursing knowledge to the patient are analyzed, the nursing effect is analyzed in combination with the recovery effect, the nursing pertinence of each keyword is obtained, the pertinence of the nursing support corresponding to the keyword is quantified, and the basis for modifying the initial index weight is provided.

[0072] Preferably, in one embodiment of the present application, please refer to Figure 3 which shows a flowchart of a nursing pertinence acquisition method provided by an embodiment of the present application, and specifically comprises:

[0073] Step S301: vectorize the patient data in the clinical nursing data, and form a binary tuple of the data vector and the recovery effect.

[0074] In order to facilitate the comparison of the differences between the clinical nursing data, the patient data is vectorized, and the recovery effect is also an important component of the clinical nursing data, so the data vector and the recovery effect form a binary tuple, which facilitates the subsequent comparison of the differences between the clinical nursing data from the two angles of the data vector of the patient and the recovery effect.

[0075] As an example, the patient data is vectorized by using one-hot encoding to obtain the data vector. The patient data includes nursing drugs, drug types used in chemotherapy treatment, incidence and composite time of oral mucositis (relative to the time after the first treatment), patient disease history and drug allergy history.

[0076] It should be noted that in other embodiments of the present application, the patient data can also be vectorized by using a bag-of-words model, a Word2Vec model, etc. The data vector can be a multi-dimensional vector, i.e. each type of data corresponds to a dimension, and when analyzing the similarity or difference between the vectors, all the same dimensions are compared, and the average of all the dimensions is taken. Both one-hot encoding and the above-mentioned technical means are well known to those skilled in the art, and will not be described in detail.

[0077] Step S302: for each keyword, in all binary tuples of the involved clinical nursing data: select the binary tuple with the largest recovery effect as the target binary tuple, and obtain the nursing targeting factor according to the difference between the recovery effect of the target binary tuple and other binary tuples, combined with the difference between the data vectors.

[0078] For nursing knowledge, if it only has special curative effect for some special patients (such as dexamethasone ointment can effectively reduce the incidence of severe oral mucositis in patients with squamous cell carcinoma), its frequency may not be high, but it has strong targeted curative effect for special patients, which is reflected in the higher recovery effect. Therefore, the binary tuple with the largest recovery effect is selected as the target binary tuple, and the difference between the target binary tuple and other binary tuples is compared.

[0079] It is considered that the greater the difference between the target binary tuple and other binary tuples in terms of recovery effect and data vector, the stronger the targeting of the nursing, so the nursing targeting factor is obtained in this way.

[0080] As an example, for each keyword, in all the binary tuples of the involved clinical nursing data, the target binary tuple is compared with each of the other binary tuples one by one, and for each of the other binary tuples: the difference between the recovery effect of the target binary tuple and the recovery effect of the other binary tuple is taken as the numerator, the cosine similarity between the data vector of the target binary tuple and the data vector of the other binary tuple is taken as the denominator, and the fraction is taken as the difference factor of the target binary tuple and the corresponding other binary tuple;

[0081] The average of the difference factors of each keyword is taken as the nursing targeting factor.

[0082] The difference between the recovery effects represents the absolute numerical difference between the recovery effects, and the cosine similarity represents the difference between the data vectors, and the smaller the cosine similarity, the greater the vector difference; meanwhile, the cosine similarity is adjusted by means of the preset correction parameter, which is added by 1 to make the value range of the cosine similarity 0-2, so as to avoid affecting the logical relationship, and is added by 0.01 to prevent the denominator from being zero; in this way, the nursing targeting factor is obtained, which reflects the difference between the clinical nursing data of a keyword and represents the nursing targeting strength.

[0083] Step S303: Fusing the nursing targeting factor and the recovery effect of the target binary tuple to obtain the nursing targeting of the corresponding keyword.

[0084] It is also considered that the greater the recovery effect of the target binary tuple, the more ideal the nursing effect, and the stronger the reference of the targeted nursing, so the nursing targeting of the corresponding keyword is finally obtained by fusing the nursing targeting factor and the recovery effect of the target binary tuple.

[0085] As an example, the product of the nursing targeting factor and the recovery effect of the target binary tuple of each keyword is taken as the nursing targeting of each keyword.

[0086] The initial index weight reflects the universality of the keyword in the historical clinical nursing record, the overall attention of the user, and the timeliness trend in the recent period, but it may not necessarily reflect the actual effectiveness of the nursing knowledge corresponding to the keyword in a specific patient group. The nursing targeting analyzes the difference in the recovery effect of the nursing knowledge on different individuals, and quantifies the curative effect strength of the nursing knowledge on specific patients. Finally, the initial index weight and the nursing targeting are fused to obtain the final index weight of each keyword, so as to balance between universality and accuracy, improve the effective hit rate of user retrieval, and improve the use efficiency and reference of the knowledge base.

[0087] Preferably, in one embodiment of the present application, when some of the keywords in the user index match the strong targeted nursing knowledge, which needs to be targeted for the special condition of the patient, these nursing knowledge should be placed at the top of the resulting form, so the product of the initial index weight of each keyword and the nursing targeting is taken as the final index weight.

[0088] In another embodiment of the present application, if the initial index weight is normalized, the nursing targeting also needs to be normalized to avoid the nursing targeting from occupying the dominant position and ignoring the initial index weight, at this time, the initial index weight and the nursing targeting can also be fused by adding or weighted summation.

[0089] It should be noted that the analysis process of the nursing targeting and the final index weight of each keyword is consistent, and only one example is described here, and the description is not repeated.

[0090] In another embodiment of the present application, after obtaining the final index weight of each keyword, the final index weight of each keyword is provided to the index module, and then the method further comprises:

[0091] Obtaining the index word input by the current user, the knowledge base system obtains a series of keyword forms matched with the index word input by the user through a string matching algorithm; arranging the keyword forms in descending order of the final index weight, and outputting the arranged keyword forms.

[0092] It should be noted that the string matching algorithm is a known technology and will not be described again; the update frequency of the final index weight of the keyword can be set to once a day, and in other embodiments of the present application, the implementer can adjust it by himself.

[0093] In summary, in view of the technical problem of low efficiency of the index mode of the existing chemotherapy-related oral mucositis nursing knowledge base, the present application provides a construction method of the chemotherapy-related oral mucositis nursing knowledge base, the present application first obtains the knowledge base and access data; further, according to the recent access trend and the total access times of each keyword, the initial index weight is obtained in combination with the distribution of the keywords in the clinical nursing data; further, according to the difference between the clinical nursing data related to each keyword, the nursing targeting is obtained in combination with the recovery effect; finally, the final index weight of each keyword is obtained by fusing the initial index weight and the nursing targeting. The scheme dynamically optimizes the index weight of the keyword by fusing the universality, timeliness and nursing targeting of the keyword, takes into account the extensive demand and individual efficacy, avoids the virtual high of hot and low efficient words, and improves the retrieval accuracy and the application value of the database.

[0094] It is to be noted that the sequential order of the above-described embodiments of the present application only for the purpose of description, but not the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0095] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the difference from other embodiments.

Claims

1. An indexing method for a knowledge base on the care of chemotherapy-related oral mucositis, characterized in that, The method includes: A knowledge base and access data were constructed by acquiring historical patients' clinical nursing data, and keywords and recovery effects were extracted from the clinical nursing data; Based on the distribution of each keyword in the clinical nursing data, a distribution weight is obtained; the access data includes the daily access count and total access count for each keyword; the recent access intensity is obtained based on the change in the daily access count for each keyword; the initial index weight for each keyword is obtained by fusing the distribution weight, the recent access intensity, and the total access count. The patient data in the clinical nursing data is vectorized, and the data vectors are combined with the recovery effects to form tuples. For each keyword, among all the tuples of the relevant clinical nursing data: the tuple with the largest recovery effect is selected as the target tuple; based on the difference in the recovery effects between the target tuple and other tuples, combined with the difference between the data vectors, a nursing targeting factor is obtained; the nursing targeting factor and the recovery effects of the target tuple are fused to obtain the nursing targeting of the corresponding keyword; the initial index weight and the nursing targeting are fused to obtain the final index weight of each keyword. The system obtains the index terms input by the current user, and uses a string matching algorithm to obtain a series of key field forms that match the user's input index terms. The key field forms are then sorted from largest to smallest according to the final index weight, and the sorted key field forms are output.

2. The indexing method for a knowledge base on chemotherapy-related oral mucositis care according to claim 1, characterized in that, The method for obtaining the distribution weights includes: The frequency of each keyword in the knowledge base is used as its respective distribution weight.

3. The indexing method for a knowledge base on chemotherapy-related oral mucositis care according to claim 1, characterized in that, The method for obtaining the recent access intensity includes: For each keyword, the recent access intensity is obtained by combining the slope of the fitted line of the number of daily visits in the preset historical neighborhood of the current day with the salient feature of the largest number of daily visits in the preset historical neighborhood of the current day.

4. The indexing method for a knowledge base on chemotherapy-related oral mucositis care according to claim 1, characterized in that, The method for obtaining the data vector includes: Patient data is vectorized using one-hot encoding to obtain data vectors.

5. The indexing method for a knowledge base on chemotherapy-related oral mucositis care according to claim 1, characterized in that, The method for obtaining the final index weight includes: The product of the initial index weight for each keyword and the nursing relevance is used as the final index weight.

6. The indexing method for a knowledge base on chemotherapy-related oral mucositis care according to claim 3, characterized in that, The preset historical neighborhood has a length of 7 days.

7. The indexing method for a knowledge base on chemotherapy-related oral mucositis care according to claim 1, characterized in that, The methods for obtaining the keywords include: The nursing knowledge question field in the preset nursing knowledge sheet is set as a key field, and the key fields involved in the clinical nursing data are identified and used as keywords.

Images