技术领域technical field
本发明属于医疗数据挖掘技术领域，具体涉及一种基于FP-Growth算法的疾病并发症挖掘方法。The invention belongs to the technical field of medical data mining, and in particular relates to a disease complication mining method based on an FP-Growth algorithm.
背景技术Background technique
数据挖掘是近年来随着人工智能和数据库技术的交叉融合而兴起的边缘学科,它致力于发现隐含在资料中的关于事物本质和事物发展趋势的知识或规律，并为专家的决策提供支持。随着信息技术在医疗行业的大规模应用，大量的医疗数据被采集起来，数据挖掘技术在医疗领域具有良好的应用前景和数据支持。从体检诊断数据库中挖掘疾病并发症以丰富专家经验和医学理论，并发症往往具有很高的复杂性和不确定性，利用海量数据研究疾病之间的并发关系进行并发症预警对疾病的治疗有重要意义。Data mining is a marginal subject emerging with the cross-integration of artificial intelligence and database technology in recent years. It is dedicated to discovering the knowledge or laws hidden in the data about the nature of things and the development trend of things, and providing support for expert decision-making. . With the large-scale application of information technology in the medical industry, a large amount of medical data is collected, and data mining technology has good application prospects and data support in the medical field. Mining disease complications from the physical examination diagnosis database to enrich expert experience and medical theory, complications are often highly complex and uncertain, using massive data to study the concurrent relationship between diseases for early warning of complications is beneficial to the treatment of diseases important meaning.
研究并发症实际上研究的是疾病之间的并发共线关系，这些疾病并发关系有些是已知的，也有些是未知的；有些是属于同一科室的，有些是跨科室的。因为其巨大的数据量，这些隐含的疾病并发关系很难为人工所发现，而数据挖掘技术正是解决这个问题的最好办法。目前的疾病并发症研究往往都只针对一种或一类疾病进行研究，例如常见的糖尿病并发症研究和某些癌症的并发症研究。The study of complications actually studies the concurrent collinear relationship between diseases. Some of these disease concurrent relationships are known and some are unknown; some belong to the same department, and some are inter-departmental. Because of its huge amount of data, it is difficult to discover these hidden concurrent disease relationships manually, and data mining technology is the best way to solve this problem. The current research on disease complications often only focuses on one or one type of disease, such as the research on common diabetes complications and the complications of certain cancers.
发明内容Contents of the invention
鉴于上述，本发明提供了一种基于FP-Growth算法的疾病并发症挖掘方法，针对所有常见疾病的并发症挖掘，旨在为医生诊断时，为患者提供更全面的体检建议以及提醒患者对一些疾病进行及早的防患。In view of the above, the present invention provides a disease complication mining method based on the FP-Growth algorithm, aiming at complication mining of all common diseases, aiming at providing more comprehensive medical examination suggestions for patients and reminding patients of some Early prevention of diseases.
一种基于FP-Growth算法的疾病并发症挖掘方法，包括如下步骤：A disease complication mining method based on FP-Growth algorithm, comprising the following steps:
(1)对医院体检数据库中的所有体检报告进行预处理以及分析，得到每份体检报告所诊断出的疾病列表；(1) Preprocessing and analyzing all the medical examination reports in the hospital medical examination database to obtain a list of diseases diagnosed by each medical examination report;
(2)基于所有体检报告所对应的疾病列表，通过统计识别输出疾病频繁项列表，该列表中的频繁项为一种疾病或两种疾病的组合，且对于任一频繁项i，其满足以下条件要求：(2) Based on the list of diseases corresponding to all physical examination reports, the list of frequent items of diseases is output through statistical identification. The frequent items in this list are a combination of one disease or two diseases, and for any frequent item i, it satisfies the following Conditional requirements:
其中：N为体检报告的总数量，support(i)为频繁项i的支持度，ρ为设定的比例阈值；Among them: N is the total number of physical examination reports, support(i) is the support degree of frequent item i, and ρ is the set ratio threshold;
(3)基于疾病频繁项列表通过计算发现关联规则，挖掘出属于频繁项的疾病所对应的并发症。(3) Based on the list of frequent items of diseases, the association rules are found through calculation, and the complications corresponding to the diseases belonging to the frequent items are mined.
所述步骤(1)中对体检报告进行预处理以及分析，具体包括对缺失值进行删除，对异常值进行处理，并统计疾病诊断的种类和分布以及生成热门疾病图，从而得到每份体检报告所诊断出的疾病列表。In the step (1), the physical examination report is preprocessed and analyzed, specifically including deleting missing values, processing abnormal values, and counting the types and distribution of disease diagnoses and generating popular disease maps, so as to obtain each physical examination report List of diseases diagnosed.
所述步骤(2)中采用FP-Growth算法统计识别出所有频繁项，从而输出疾病频繁项列表。In the step (2), the FP-Growth algorithm is used to statistically identify all frequent items, thereby outputting a list of frequent disease items.
所述支持度support(i)为疾病列表中包含频繁项i的体检报告数量。The support degree support(i) is the number of medical examination reports containing frequent item i in the disease list.
所述步骤(3)的具体实现过程如下：The concrete realization process of described step (3) is as follows:
3.1对于属于频繁项的任一疾病a，统计与其组合成频繁项的所有关联疾病；3.1 For any disease a belonging to a frequent item, count all associated diseases combined with it to form a frequent item;
3.2对于疾病a的任一关联疾病b，通过以下算式计算两者的可信度confidence(a/b)：3.2 For any associated disease b of disease a, the confidence (a/b) of the two is calculated by the following formula:
其中：support(a)为仅由疾病a组成的频繁项的支持度，即疾病列表中包含该频繁项的体检报告数量；support(a/b)为由疾病a和关联疾病b组合的频繁项的支持度，即疾病列表中包含该频繁项的体检报告数量；Among them: support(a) is the support degree of the frequent item consisting only of disease a, that is, the number of medical examination reports containing the frequent item in the disease list; support(a/b) is the frequent item combined by disease a and associated disease b The support degree of , that is, the number of medical examination reports containing the frequent item in the disease list;
3.3判断可信度confidence(a/b)是否大于预设的可信度阈值，若是，则判定关联疾病b为疾病a的并发症；3.3 Determine whether the reliability confidence (a/b) is greater than the preset reliability threshold, if so, determine that the associated disease b is a complication of disease a;
3.4根据步骤3.2～3.3遍历疾病a的所有关联疾病，挖掘得到疾病a的所有并发症，进而根据可信度confidence(a/b)对这些并发症降序排列后展示。3.4 Traverse all associated diseases of disease a according to steps 3.2 to 3.3, dig out all complications of disease a, and then display these complications in descending order according to confidence (a/b).
本发明疾病并发症挖掘方法基于大型医院多年间的体检数据，对患者的诊断数据进行提取，并利用FP-Growth算法得到频繁项集，从中构造可信度不低于阈值的规则，即疾病并发症。医生在给出诊断建议时，不仅可以根据患者体检数据进行建议，还可以根据疾病并发症对患者提出科学可靠的建议和防患措施。由此，本发明具有以下有益技术效果：The disease complication mining method of the present invention is based on the physical examination data of large hospitals for many years, extracts the patient's diagnostic data, and uses the FP-Growth algorithm to obtain frequent itemsets, and constructs a rule whose reliability is not lower than the threshold, that is, the disease complication disease. When giving diagnostic advice, doctors can not only make recommendations based on the patient's physical examination data, but also provide scientific and reliable advice and preventive measures to patients based on disease complications. Thus, the present invention has the following beneficial technical effects:
(1)本发明数据来源于大型医院多年间的体检数据，诊断数据多达50多万条，通过关联规则挖掘得到的疾病关联症全面、真实、可靠。(1) The data of the present invention comes from the physical examination data of large hospitals for many years, and the diagnostic data reaches more than 500,000 pieces. The disease-related diseases obtained through association rule mining are comprehensive, true and reliable.
(2)医院体检数据每天都在增加，本发明可以设置更新时间，使疾病并发症数据保持相对实时性。(2) The hospital physical examination data is increasing every day, and the present invention can set the update time to keep the disease complication data relatively real-time.
(3)本发明采用FP-Growth算法，比一般被采用的Apriori关联规则算法更快速、高效。(3) The present invention adopts the FP-Growth algorithm, which is faster and more efficient than the generally adopted Apriori association rule algorithm.
(4)本发明除了给出疾病的并发症，还给出相应的可能性，并按照可能性高低对并发症进行排序，使提供给病患的诊断结果和诊断建议更加准确，提高病患体检满意度。(4) In addition to giving the complications of the disease, the present invention also gives the corresponding possibility, and sorts the complications according to the possibility, so that the diagnostic results and diagnostic suggestions provided to the patient are more accurate, and the medical examination of the patient is improved. satisfaction.
附图说明Description of drawings
图1为本发明疾病并发症挖掘方法的流程示意图。Fig. 1 is a schematic flowchart of the method for mining disease complications of the present invention.
图2为本发明中数据预处理及分析部分的流程示意图。Fig. 2 is a schematic flow chart of the data preprocessing and analysis part in the present invention.
图3为常见疾病诊断示意图。Figure 3 is a schematic diagram of the diagnosis of common diseases.
图4为本发明中识别频繁项的流程示意图。Fig. 4 is a schematic flow chart of identifying frequent items in the present invention.
图5为疾病诊断得到的并发症展示图。Fig. 5 is a display diagram of complications obtained from disease diagnosis.
具体实施方式detailed description
为了更为具体地描述本发明，下面结合附图及具体实施方式对本发明的技术方案进行详细说明。In order to describe the present invention more specifically, the technical solutions of the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.
本实施方式基于浙江大学医学院附属第二医院的十年间的体检数据，对患者的诊断数据进行提取，并利用FP-Growth算法得到频繁项集，从中构造可信度不低于阈值的规则，即疾病并发症。医生在给出诊断建议时，不仅可以根据患者体检数据进行建议，还可以根据疾病并发症对患者提出科学可靠的建议和防患措施。整个挖掘方法主要由三个部分组成：数据预处理和分析、识别频繁项集、发现关联规则。This embodiment is based on the physical examination data of the Second Affiliated Hospital of Zhejiang University School of Medicine for ten years, extracts the patient's diagnostic data, and uses the FP-Growth algorithm to obtain frequent itemsets, and constructs rules with a reliability not lower than the threshold. a disease complication. When giving diagnostic advice, doctors can not only make recommendations based on the patient's physical examination data, but also provide scientific and reliable advice and preventive measures to patients based on disease complications. The whole mining method mainly consists of three parts: data preprocessing and analysis, identification of frequent itemsets, and discovery of association rules.
本数据为体检诊断数据，具有不完整性、冗余性以及格式多变性。所以在数据预处理部分主要对缺失值进行删除，对异常值进行处理，并统计疾病诊断的种类和分布以及生成热门疾病图。This data is physical examination diagnosis data, which is incomplete, redundant and variable in format. Therefore, in the data preprocessing part, the missing values are mainly deleted, the outliers are processed, and the types and distributions of disease diagnoses are counted, and popular disease maps are generated.
识别频繁项集部分利用FP-Growth算法识别所有的疾病诊断频繁项集，要求频繁项目集的支持率不低于设定的最低值。此部分是发现疾病并发症的关键部分，也是计算最大的部分。The part of identifying frequent itemsets uses the FP-Growth algorithm to identify all frequent itemsets for disease diagnosis, and requires that the support rate of frequent itemsets is not lower than the set minimum value. This part is the key part of discovering the complication of the disease, and also the part with the largest calculation.
发现关联规则部分是从频繁项目集中构造可信度不低于用户设定的最低值的规则，并用数据可视化工具直观展示出来。The part of discovering association rules is to construct rules whose credibility is not lower than the minimum value set by the user from the frequent item set, and display them intuitively with data visualization tools.
如图1所示，本实施方式先对数据库中数据进行预处理并抽取生成诊断数据集，然后利用FP-Growth算法生成频繁项集，其中频繁项集的支持度大于设定的最低支持度，再利用频繁项集进行关联规则挖掘，找到高于设定的最低可信度的关联规则集，就是疾病并发症数据。As shown in Figure 1, this embodiment first preprocesses the data in the database and extracts it to generate a diagnostic data set, and then uses the FP-Growth algorithm to generate frequent itemsets, where the support degree of the frequent itemsets is greater than the set minimum support degree, Then use frequent itemsets to mine association rules, and find association rule sets higher than the set minimum reliability, which is the disease complication data.
图2为数据预处理和分析模块的流程，数据存储在Oracle服务器中的体检诊断信息表中，需要用得到的键值为体检编码和诊断信息，一次体检对应一个体检编码，一次体检有多个诊断信息。先把诊断信息中类似“：+-”和“？”这样的无意义的字符过滤掉，之后再删除诊断信息中的缺失值和冗余值；把数据从数据库中提取出来之后统计疾病诊断分布情况并重新整合成频繁项集挖掘所需格式。疾病诊断种类约为12万种，常见疾病诊断如图3所示，字体越大代表此疾病诊断在体检中出现的次数越多。Figure 2 shows the flow of the data preprocessing and analysis module. The data is stored in the physical examination diagnosis information table in the Oracle server. The key values that need to be obtained are the physical examination code and diagnostic information. One physical examination corresponds to one physical examination code, and one physical examination has multiple diagnostic information. First filter out meaningless characters like ":+-" and "?" in the diagnostic information, and then delete the missing and redundant values in the diagnostic information; after extracting the data from the database, count the disease diagnosis distribution situation and reintegrated into the required format for frequent itemset mining. There are about 120,000 types of disease diagnoses. Common disease diagnoses are shown in Figure 3. The larger the font size, the more times the disease diagnosis appears in the physical examination.
图4为频繁项生成过程，采用FP-Growth算法找出满足最小支持度的所有频繁项集。FP-Growth算法采用分而治之策略：将提供频繁项目集的事务数据库压缩到一颗频繁模式树(FP-tree)，但仍保留项目集关联信息；然后将这种压缩后的数据库分成一组条件数据库，每个关联一个频繁项目，并分别挖掘每个条件数据库。本实施方式只需得到长度为1和长度为2的频繁项，所以对循环条件做些改变。Figure 4 shows the frequent item generation process, using the FP-Growth algorithm to find all frequent item sets that meet the minimum support. The FP-Growth algorithm adopts a divide-and-conquer strategy: compress the transaction database that provides frequent item sets into a frequent pattern tree (FP-tree), but still retain the item set association information; then divide this compressed database into a set of conditional databases , each associated with a frequent item, and mining each conditional database separately. This embodiment only needs to obtain frequent items with length 1 and length 2, so some changes are made to the loop conditions.
算法：FP-Growth//使用FP-tree通过模式段增长，挖掘频繁模式。Algorithm: FP-Growth//Use FP-tree to grow through pattern segments to mine frequent patterns.
输入：事务数据库D，最小支持度阈值min_sup。Input: transaction database D, minimum support threshold min_sup.
输出：频繁模式的完全集。Output: Complete set of frequent patterns.
方法：method:
(1)按以下步骤构造FP-tree：(1) Construct FP-tree according to the following steps:
(a)扫描事务数据库D一次。收集频繁项的集合F和它们的支持度。对F按支持度降序排序，结果为频繁项集L。(a) Scan the transactional database D once. Collect the set F of frequent items and their support. Sort F in descending order of support, and the result is frequent itemset L.
(b)创建FP-tree的根节点，以“null”标记它。对于D中每个事务Trans，执行：(b) Create the root node of the FP-tree and mark it with "null". For each transaction Trans in D, execute:
选择Trans中的频繁项,并按L中的次序排序。设排序后的频繁项表为[p|P],其中p是第一个元素，而P是剩余元素的表。调用insert_tree([p|P]，T)。该过程执行情况如下。如果T有子女N使得N.item-name＝p.item-name，则N的计数增加1；否则创建一个新节点N，将其计数设置为1，链接到它的父节点T，并且通过节点链结构将其链接到具有相同item-name的节点。如果P非空，递归地调用insert_tree(P，N)。Select the frequent items in Trans and sort them in the order of L. Let the sorted frequent item list be [p|P], where p is the first element, and P is the list of remaining elements. Call insert_tree([p|P], T). The process is performed as follows. If T has children N such that N.item-name = p.item-name, then increment the count of N by 1; otherwise create a new node N, set its count to 1, link to its parent T, and pass the node A chain structure links it to nodes with the same item-name. If P is non-null, call insert_tree(P, N) recursively.
(2)FP-tree的挖掘通过调用过程FP-Growth(FP-tree,null)实现。(2) The mining of FP-tree is realized by calling the procedure FP-Growth(FP-tree, null).
//该过程实现如下：//The process is implemented as follows:
Procedure FP-Growth(tree,α)Procedure FP-Growth(tree,α)
1)if tree包含单个路径P then1) if tree contains a single path P then
2)for路径P的每个节点组合(记为β)2) Each node combination of the for path P (denoted as β)
3)产生模式β∪α，支持度support＝β中节点的最小支持度3) Generate pattern β∪α, support degree support=minimum support degree of nodes in β
4)else for each ai在Tree的头部{4) else for each ai at the head of Tree {
5)产生模式β＝ai∪β，其支持度support＝ai·support5) Generate pattern β＝ai ∪β, and its support degree support＝ai ·support
6)构造模式β的条件模式基，并构造β的条件FP-treeβ6) Construct the conditional pattern base of pattern β, and construct the conditional FP-tree β ofβ
7)ifthen7)if then
8)调用FP-Growth(treeβ,β)}8) Call FP-Growth(treeβ , β)}
得到频繁项集后，根据量化指标可信度得到关联规则。一条规则P→H的可信度经计算方式如下式：After obtaining frequent itemsets, association rules are obtained according to the credibility of quantitative indicators. The credibility of a rule P→H is calculated as follows:
confidence(P→H)＝support(P|H)/support(P)confidence(P→H)＝support(P|H)/support(P)
其中：P|H是指所有出现在集合P或者H中的元素。同样因项目需求，P和H的长度设置为1。疾病并发症数据就是求所有满足最小可信度的规则集合，同时，同一疾病的不同并发症按照可信度进行降序排序，最后得到的疾病并发症如图5所示，线条相连的两个疾病代表互为并发症。Among them: P|H refers to all elements that appear in the set P or H. Also due to project requirements, the lengths of P and H are set to 1. The data of disease complications is to find all the rule sets that satisfy the minimum reliability. At the same time, different complications of the same disease are sorted in descending order according to the reliability. The final disease complications are shown in Figure 5. Two diseases connected by lines Represents each other as complications.
上述对实施例的描述是为便于本技术领域的普通技术人员能理解和应用本发明。熟悉本领域技术的人员显然可以容易地对上述实施例做出各种修改，并把在此说明的一般原理应用到其他实施例中而不必经过创造性的劳动。因此，本发明不限于上述实施例，本领域技术人员根据本发明的揭示，对于本发明做出的改进和修改都应该在本发明的保护范围之内。The above description of the embodiments is for those of ordinary skill in the art to understand and apply the present invention. It is obvious that those skilled in the art can easily make various modifications to the above-mentioned embodiments, and apply the general principles described here to other embodiments without creative efforts. Therefore, the present invention is not limited to the above embodiments, and improvements and modifications made by those skilled in the art according to the disclosure of the present invention should fall within the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611168316.7ACN106709248A (en) | 2016-12-16 | 2016-12-16 | Disease complication excavating method based on FP-Growth algorithm |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611168316.7ACN106709248A (en) | 2016-12-16 | 2016-12-16 | Disease complication excavating method based on FP-Growth algorithm |
| Publication Number | Publication Date |
|---|---|
| CN106709248Atrue CN106709248A (en) | 2017-05-24 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611168316.7APendingCN106709248A (en) | 2016-12-16 | 2016-12-16 | Disease complication excavating method based on FP-Growth algorithm |
| Country | Link |
|---|---|
| CN (1) | CN106709248A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451416A (en)* | 2017-08-28 | 2017-12-08 | 昆明理工大学 | A kind of sle auxiliary diagnostic equipment and method |
| CN109147879A (en)* | 2018-07-02 | 2019-01-04 | 北京众信易保科技有限公司 | The method and system of Visual Report Forms based on medical document |
| CN110019188A (en)* | 2017-09-15 | 2019-07-16 | 上海诺悦智能科技有限公司 | A kind of suspicious characteristic discovery method based on trade network node |
| CN111785372A (en)* | 2020-05-14 | 2020-10-16 | 浙江知盛科技集团有限公司 | Collaborative filtering disease prediction system based on association rule and electronic equipment thereof |
| CN113643815A (en)* | 2021-08-31 | 2021-11-12 | 平安医疗健康管理股份有限公司 | Prediction method, device, computer equipment and storage medium for disease complications |
| CN113823414A (en)* | 2021-08-23 | 2021-12-21 | 杭州火树科技有限公司 | Main diagnosis and main operation matching detection method and device, computing equipment and storage medium |
| CN115953254A (en)* | 2022-12-30 | 2023-04-11 | 杭州火树科技有限公司 | A disease course chain identification method and device |
| CN118197591A (en)* | 2022-12-12 | 2024-06-14 | 于清 | Information processing methods applied to big data smart healthcare |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451416A (en)* | 2017-08-28 | 2017-12-08 | 昆明理工大学 | A kind of sle auxiliary diagnostic equipment and method |
| CN110019188A (en)* | 2017-09-15 | 2019-07-16 | 上海诺悦智能科技有限公司 | A kind of suspicious characteristic discovery method based on trade network node |
| CN109147879A (en)* | 2018-07-02 | 2019-01-04 | 北京众信易保科技有限公司 | The method and system of Visual Report Forms based on medical document |
| CN109147879B (en)* | 2018-07-02 | 2021-07-27 | 北京众信易保科技有限公司 | Method and system for visual report based on medical document |
| CN111785372A (en)* | 2020-05-14 | 2020-10-16 | 浙江知盛科技集团有限公司 | Collaborative filtering disease prediction system based on association rule and electronic equipment thereof |
| CN113823414A (en)* | 2021-08-23 | 2021-12-21 | 杭州火树科技有限公司 | Main diagnosis and main operation matching detection method and device, computing equipment and storage medium |
| CN113823414B (en)* | 2021-08-23 | 2024-04-05 | 杭州火树科技有限公司 | Main diagnosis and main operation matching detection method, device, computing equipment and storage medium |
| CN113643815A (en)* | 2021-08-31 | 2021-11-12 | 平安医疗健康管理股份有限公司 | Prediction method, device, computer equipment and storage medium for disease complications |
| CN118197591A (en)* | 2022-12-12 | 2024-06-14 | 于清 | Information processing methods applied to big data smart healthcare |
| CN115953254A (en)* | 2022-12-30 | 2023-04-11 | 杭州火树科技有限公司 | A disease course chain identification method and device |
| CN115953254B (en)* | 2022-12-30 | 2025-07-04 | 杭州火树科技有限公司 | A disease course chain identification method and device |
| Publication | Publication Date | Title |
|---|---|---|
| CN106709248A (en) | Disease complication excavating method based on FP-Growth algorithm | |
| Shibata et al. | Detecting emerging research fronts based on topological measures in citation networks of scientific publications | |
| Huang et al. | A likelihood ratio test based method for signal detection with application to FDA’s drug safety data | |
| Wongsuphasawat et al. | Outflow: Visualizing patient flow by symptoms and outcome | |
| CN107833595A (en) | Medical big data multicenter integration platform and method | |
| Fryback et al. | Bayesian cost-effectiveness analysis: an example using the GUSTO trial | |
| CN106250288A (en) | A Root Alarm Analysis and Recognition Method Based on Data Mining | |
| Fernandez-Basso et al. | A fuzzy-based medical system for pattern mining in a distributed environment: Application to diagnostic and co-morbidity | |
| US11145394B2 (en) | Clinical quality analytics system with recursive, time sensitive event-based protocol matching | |
| CN104573306A (en) | Method and system for visual analysis of clinical episodes | |
| CN110277172A (en) | A clinical medication behavior analysis system based on efficient negative sequence mining mode and its working method | |
| CN113362960A (en) | Urban resident public health influence factor visual analysis system and method combining multi-source data | |
| CN118824567A (en) | A medical diagnosis data fusion analysis method and system based on decision tree model | |
| CN117831701A (en) | A quality control method for electronic medical records based on rule engine | |
| Mahammad et al. | Machine learning approach to predict asthma prevalence with decision trees | |
| CN111028088A (en) | Group cheating and insurance behavior identification method and system based on frequent set mining | |
| WO2022088674A1 (en) | Essential computing-oriented differential content recommendation method across data, information and knowledge modalities | |
| Lu et al. | Timeline and episode-structured clinical data: Pre-processing for Data Mining and analytics | |
| Patil et al. | Predicting burn patient survivability using decision tree in weka environment | |
| Bakr et al. | Cooperative spatial decision support system for controlling animal diseases outbreaks in Egypt | |
| Aiswarya et al. | Detection and prediction of frequent diseases in India through association technique using apriori algorithm and random forest regression | |
| Hryhoruk et al. | Mining Sequential Patterns with Timelines from Digital Health Data | |
| Fang et al. | Abnormal event health-status monitoring based on multi-dimensional and multi-level association rules constraints in nursing information system | |
| Rao et al. | Extracting Insights and Prognosis of Corona Disease | |
| CN115511683B (en) | Public health data acquisition and processing system |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20170524 | |
| RJ01 | Rejection of invention patent application after publication |