CN107766881B - Way finding method and device based on basic classifier and storage device - Google Patents
Way finding method and device based on basic classifier and storage device Download PDFInfo
- Publication number
- CN107766881B CN107766881B CN201710944786.6A CN201710944786A CN107766881B CN 107766881 B CN107766881 B CN 107766881B CN 201710944786 A CN201710944786 A CN 201710944786A CN 107766881 B CN107766881 B CN 107766881B
- Authority
- CN
- China
- Prior art keywords
- classifier
- basic
- knn
- bayes
- data
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention provides a way-finding method, a way-finding device and a storage device based on a basic classifier, wherein the method comprises the following steps: establishing a database to record environmental data; extracting expected geographic location data; determining a base classifier with improved base classifier parameters using the environmental data and expected geographic location data; integrating the improved basic classifier to obtain a strong classifier; and carrying out path finding by using the strong classifier and obtaining a path planning result. A way-finding device and a storage device based on a basic classifier are used for realizing the way-finding method based on the basic classifier. The invention effectively improves the classification accuracy of the classifier, and further can provide a better result for route prediction.
Description
Technical Field
The invention relates to the field of data mining, in particular to a way finding method, a way finding device and a way finding storage device based on a basic classifier.
Background
Geo-tagging refers to the application of geospatial metadata (i.e., latitude, longitude) to a series of web-based media (e.g., photos, videos, articles). With the development of mobile internet and social networks and the popularization of intelligent terminals with built-in GPS devices, a great number of photos and videos with geographic tags are generated on the networks. The geo-tagging of these photos and videos provides a wealth of information about user behavior, the potential of which is increasing. And the rich location-based data provided by geotagged photos and videos can potentially be used to provide some specific location information and services. Recently, information of these geotagged photographs has been increasingly used for travel recommendations. Based on this basis, how to classify the geographic information quickly and efficiently and generate an accurate and effective route becomes a current concern.
Disclosure of Invention
The invention provides a path finding method, device and storage device based on a basic classifier, wherein the basic classifier is integrated to obtain a strong classifier, so that the classification accuracy of the classifier is effectively improved, and the problem is effectively solved.
The technical scheme provided by the invention is as follows: a way-finding method based on a basic classifier, the method comprising the steps of: establishing a database to record environmental data; extracting expected geographic location data; determining a base classifier with improved base classifier parameters using the environmental data and expected geographic location data; integrating the improved basic classifier to obtain a strong classifier; and carrying out path finding by using the strong classifier and obtaining a path planning result. A storage device storing instructions and data for implementing the base classifier-based way-finding method. A way-finding apparatus based on a basic classifier, the apparatus comprising a processor and the storage device; the processor loads and executes instructions and data in the storage device for realizing the basic classifier-based way-finding method.
The invention has the beneficial effects that: the invention provides a path finding method, device and storage device based on a basic classifier, wherein the basic classifier is integrated to obtain a strong classifier, so that the classification accuracy of the classifier is effectively improved, and a higher-quality result can be provided for route prediction.
Drawings
FIG. 1 is a general flow chart of a way-finding method based on a basic classifier according to an embodiment of the present invention;
FIG. 2 is a flow chart of the steps of improving the BAyes classifier in the embodiment of the present invention;
FIG. 3 is a flow chart of the steps of an embodiment of the present invention in which the Knn classifier is modified;
FIG. 4 is a flow chart of the path finding step using the final BAyes-Knn strong classifier in the embodiment of the present invention;
fig. 5 is a schematic diagram of the operation of the hardware device according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, embodiments of the present invention will be further described with reference to the accompanying drawings, wherein specific technical details are set forth below, such as: methods, apparatus, etc. may be used solely for the purpose of facilitating a better understanding of the subject matter and are not intended to limit the invention to the details shown.
The embodiment of the invention provides a way finding method and device based on a basic classifier and a storage device, which are used for solving the problem that the way finding method is not suitable for the way finding device. Referring to fig. 1, fig. 1 is an overall flowchart of a basic classifier-based routing method in an embodiment of the present invention, where the method is implemented by a basic classifier-based routing device, and includes the specific steps of:
s101: establishing a database to record environmental data, wherein the environmental data can influence the trip of a user and specifically comprises the following steps: weather, temperature, holidays, and light and vigorous seasons.
S102: extracting expected geographical position data, wherein the expected geographical position is a hot tourist attraction; the popular tourist attractions are obtained by performing spatial clustering analysis and label semantic information extraction on a large amount of photo data through a CFSFDP-based DBSCAN algorithm and a TF-IDF algorithm, and specifically comprise the following steps: and mapping the photo data to the corresponding hot scenery spots.
S103: determining base classifier parameters using the environmental data and the desired geographic location data results in improved base classifiers, which are the BAyes classifier and the Knn classifier.
S104: and integrating the improved basic classifier to obtain a strong classifier, specifically, weighting and integrating the BAyes classifier and the Knn classifier by using an Adaboost algorithm to obtain a BAyes-Knn strong classifier, and optimizing the BAyes-Knn strong classifier by using a 10-fold cross-validation method. The strong classifier is used for path finding, i.e. determining the optimal route. Testing the BAyes-Knn strong classifier to obtain a final BAyes-Knn strong classifier, which specifically comprises the following steps: determining an evaluation criterion for the test comprising: number consistency, sequence consistency, data integrity, and route overlap. The formula for each evaluation criterion is as follows:
q and P are the node numbers of the actual value line and the predicted value line ', and the node numbers appearing in the line and the line' at the same time are set as H; q (Q-1) and P (P-1) node ordered pairs can be decomposed together in the line and the line ', and the number of the ordered pairs which are simultaneously present in the line and the line' is set as M. CR is a comprehensive decision value of DI, OC and NC, in order to ensure that the maximum value of CR is 1, the denominator is set to be 2, and the larger the CR value is, the higher the coincidence degree of the predicted value and the actual value is, and the more accurate the recommendation result of the classifier is.
S105: and carrying out path finding by using the strong classifier and obtaining a path planning result.
Referring to fig. 2, fig. 2 is a flowchart of steps of improving a BAyes classifier in the embodiment of the present invention, and the specific steps include:
s201: and combining the conditional attributes with strong correlation to reduce the error caused by the conditional independence assumption.
S202: a kernel function is introduced and the bandwidth is determined based on the conditional attribute weights. And the prior probability of the condition attribute is smoothly estimated by utilizing the kernel function, so that the error is reduced. The specific calculation formula is as follows:
wherein, WiRepresenting the bandwidth, n is the dimension of the training data set,representing a prior probability As a priori probabilityAn estimate of (d).
Referring to fig. 3, fig. 3 is a flowchart of steps of improving the Knn classifier in the embodiment of the present invention, and the specific steps include:
s301: and weighting the condition attribute. Since different conditional attributes have different degrees of influence on the final decision, the different conditional attributes are weighted by mutual information and correlation coefficients, and the weighting formula is as follows:
wherein p is the number of conditional attributes, I (C)i,Sj) Represents a condition attribute CiMutual information with decision attribute S, Cov (C)iS) represents CiCovariance with S, D (C)i) And D (S) respectively represent CiAnd the variance of S.
S302: and calculating a user similarity matrix. And carrying out personalized setting on the users by utilizing the similarity among the users, wherein the greater the similarity is, the more similar the hobbies are.
S303: and calculating the actual spatial distance matrix of the scenic spots by utilizing the spatial distribution of the hot scenic spots.
S304: and the similarity calculation method is improved, and each item in the traditional similarity formula is weighted by using the condition attribute weight and the spatial distance between the scenic spots. The formula used for weighting is:
wherein t and d are two different user trip records, Dis (t, d) represents the distance between t and d, p is the number of condition attributes, sim (t, d) represents the similarity between t and d, sim _ uusert,userd is the user attribute user through t and dt、,userdIn the user similarity matrix UsimThe found corresponding similarity value is used for user preference setting. The similarity matrix UsimComprises the following steps:
where sim _ uijRepresenting the similarity of users i and j, and each row represents the similarity vector of one user with all other users.
After all the similarities are calculated, voting is carried out by utilizing the average similarity. Suppose that among K neighbors, the decision value is SiThe training data of (a) has0,d1...dq(q < K), then SiThe average similarity calculation method of (2) is as follows:
s305: the number of neighbors is determined. And calculating the average accuracy rate change trend of the basic KNN classifier under different K values by adopting a 10-fold cross verification method to determine the final K value.
Referring to table 1, table 1 is a specific definition of all relevant parameters of the formulas in the above-mentioned BAyes classifier and knn classifier improvement steps.
TABLE 1
Referring to fig. 4, fig. 4 is a flowchart of the step of performing the path finding by using the final BAyes-Knn strong classifier in the embodiment of the present invention, and the specific steps include:
s401: the start and end points of the route are determined.
S402: predicting the first tourist attraction S to arrive from a starting point1。
S403: according to S1Predicting the next tourist attraction S2。
S404: and judging whether the predicted tourist attractions are repeated.
S405: if so, the MDW score sequence output by the BAyes-Knn strong classifier is used to select a tourist attraction having a score next to the predicted tourist attraction.
S406: if not, the steps of predicting the tourist attractions are continuously executed until the predicted tourist attractions are the destination.
S407: and arranging the predicted tourist attractions into a sequence to obtain a final searched route.
Referring to fig. 5, fig. 5 is a schematic diagram of a hardware device according to an embodiment of the present invention, where the hardware device specifically includes: a way-finding device 501 based on a basic classifier, a processor 502 and a storage device 503.
The processor 502: the processor 502 loads and executes the instructions and data in the storage device 503 to implement a basic classifier-based way-finding method.
The storage device 503: the storage device 503 stores instructions and data; the storage device 503 is used to implement the basic classifier-based way-finding method.
All the technical features of the claims of the present invention are elaborated upon by implementing the embodiments of the present invention.
Different from the prior art, the embodiment of the invention provides a path finding method, device and storage device based on a basic classifier, and a strong classifier is obtained by integrating the basic classifier, so that the classification accuracy of the classifier is effectively improved, and a higher-quality result can be provided for route prediction.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (7)
1. A routing method based on a basic classifier is realized by hardware equipment, and is characterized in that: the method comprises the following steps: establishing a database to record environmental data; extracting expected geographic location data; determining a base classifier with improved base classifier parameters using the environmental data and expected geographic location data; integrating the improved basic classifier to obtain a strong classifier; carrying out path finding by using the strong classifier and obtaining a path planning result;
the extracting of the expected geographic position data is specifically as follows: the expected geographical position is a hot tourist attraction; the hot scenic spots are obtained by performing spatial clustering analysis and label semantic information extraction on a large amount of photo data through a CFSFDP-based DBSCAN algorithm and a TF-IDF algorithm, namely, the photo data are mapped to the corresponding hot scenic spots;
the determining the improved base classifier parameters using the environmental data and expected geographic location data to obtain an improved base classifier comprises: the basic classifiers are a BAyes classifier and an Knn classifier; the improvement steps of the BAyes classifier comprise: combining the conditional attributes with strong correlation to weaken the error caused by conditional independence assumption; introducing a kernel function, and determining the bandwidth based on the condition attribute weight; knn the improvement steps of the classifier include: weighting the condition attribute; calculating a user similarity matrix; calculating an actual spatial distance matrix of the scenic spots by utilizing the spatial distribution of the popular tourist attractions; the similarity calculation method is improved, and each item in the traditional similarity formula is weighted by using the condition attribute weight and the spatial distance between the scenic spots; the number of neighbors is determined.
2. The method of claim 1, wherein the method comprises: the environmental data can influence the user trip, specifically include: weather, temperature, holidays, and light and vigorous seasons.
3. The method of claim 1, wherein the method comprises: the method for obtaining the strong classifier by integrating the improved basic classifier specifically comprises the following steps: and (3) carrying out weighted integration on the BAyes classifier and the Knn classifier by using an Adaboost algorithm to obtain a BAyes-Knn strong classifier, and optimizing the BAyes-Knn strong classifier by using a 10-fold cross-validation method.
4. A way-finding method based on a basic classifier as claimed in claim 3, characterized in that: testing the BAyes-Knn strong classifier to obtain a final BAyes-Knn strong classifier, which specifically comprises the following steps: determining an evaluation criterion for the test comprising: number consistency, sequence consistency, data integrity, and route overlap.
5. The method of claim 4, wherein the path-finding method based on the basic classifier comprises: and carrying out path finding by using the final BAyes-Knn strong classifier, and specifically comprising the following steps: determining a starting point and an end point of a route; predicting the first tourist attraction S to arrive from a starting point1(ii) a According to S1Predicting the next tourist attraction S2(ii) a Judging whether the predicted tourist attractions are repeated or not; if so, the MDW score sequence output by the BAyes-Knn strong classifier is used to select a score next to the predicted travelTourist attractions of the attraction; if not, continuing to execute the steps of predicting the tourist attractions until the predicted tourist attractions are the destination; and arranging the predicted tourist attractions into a sequence to obtain a final searched route.
6. A storage device, comprising: the storage device stores instructions and data for implementing any one of the way-finding methods based on the basic classifier as claimed in claims 1-5.
7. A seek way equipment based on basic classifier which characterized in that: the method comprises the following steps: a processor and the storage device; the processor loads and executes the instructions and data in the storage device in claim 6 to realize any one of the way-finding methods based on the basic classifier in claims 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710944786.6A CN107766881B (en) | 2017-09-30 | 2017-09-30 | Way finding method and device based on basic classifier and storage device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710944786.6A CN107766881B (en) | 2017-09-30 | 2017-09-30 | Way finding method and device based on basic classifier and storage device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107766881A CN107766881A (en) | 2018-03-06 |
CN107766881B true CN107766881B (en) | 2020-06-26 |
Family
ID=61267075
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710944786.6A Active CN107766881B (en) | 2017-09-30 | 2017-09-30 | Way finding method and device based on basic classifier and storage device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107766881B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110795524B (en) * | 2019-10-31 | 2022-07-05 | 望海康信(北京)科技股份公司 | Main data mapping processing method and device, computer equipment and storage medium |
CN113314209B (en) * | 2021-06-11 | 2023-04-18 | 吉林大学 | Human body intention identification method based on weighted KNN |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102278995A (en) * | 2011-04-27 | 2011-12-14 | 中国石油大学(华东) | Bayes path planning device and method based on GPS (Global Positioning System) detection |
CN103512581A (en) * | 2012-06-28 | 2014-01-15 | 北京搜狗科技发展有限公司 | Path planning method and device |
CN106408015A (en) * | 2016-09-13 | 2017-02-15 | 电子科技大学成都研究院 | Road fork identification and depth estimation method based on convolutional neural network |
-
2017
- 2017-09-30 CN CN201710944786.6A patent/CN107766881B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102278995A (en) * | 2011-04-27 | 2011-12-14 | 中国石油大学(华东) | Bayes path planning device and method based on GPS (Global Positioning System) detection |
CN103512581A (en) * | 2012-06-28 | 2014-01-15 | 北京搜狗科技发展有限公司 | Path planning method and device |
CN106408015A (en) * | 2016-09-13 | 2017-02-15 | 电子科技大学成都研究院 | Road fork identification and depth estimation method based on convolutional neural network |
Non-Patent Citations (6)
Title |
---|
Comprehensive Predictions of Tourists’ Next Visit Location Based on Call Detail Records using Machine Learning and Deep Learning methods;Naichun Chen et al.;《2017 IEEE 6th International Congress on Big Data》;20170911;第1-6页 * |
Personalized and Situation-Aware Multimodal Route Recommendations: The FAVOUR Algorithm;Paolo Campigotto et al.;《IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS》;20170131;第18卷(第1期);第92,95-97页 * |
Planning for tourism routes using social networks;Isabel Cenamor et al.;《Expert Systems With Applications 》;20161019;第1-9页 * |
Recommending research collaborations using link prediction and random forest classifiers;Raf Guns et al.;《 Scientometrics》;20141231;第101卷(第2期);第1-14页 * |
Travel Recommendation by Mining People Attributes and Travel Group Types From Community-Contributed Photos;Yan-Ying Chen et al.;《IEEE TRANSACTIONS ON MULTIMEDIA》;20131031;第15卷(第6期);第1287页 * |
基于地理标记照片的大西安旅游圈游客时空行为研究;张少杰;《中国优秀硕士学位论文全文数据库 经济与管理科学辑》;20160315;第2016年卷(第3期);第J153-144页 * |
Also Published As
Publication number | Publication date |
---|---|
CN107766881A (en) | 2018-03-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107291945B (en) | High-precision clothing image retrieval method and system based on visual attention model | |
TWI619030B (en) | Method and device for pushing track information | |
Cao et al. | Aworldwide tourism recommendation system based on geotaggedweb photos | |
US9123259B2 (en) | Discovering functional groups of an area | |
US11861516B2 (en) | Methods and system for associating locations with annotations | |
US20110184949A1 (en) | Recommending places to visit | |
CN102591868B (en) | System and method for automatic generation of photograph guide | |
US10769428B2 (en) | On-device image recognition | |
US20180340787A1 (en) | Vehicle routing guidance to an authoritative location for a point of interest | |
US20140222783A1 (en) | Systems and methods for automatically determining an improved view for a visual query in a mobile search | |
CN104520848B (en) | According to attendant's search events | |
CN112214677B (en) | Point of interest recommendation method and device, electronic equipment and storage medium | |
US20230244736A1 (en) | Generating Location Based Photo Discovery Suggestions | |
EP3274943A1 (en) | Systems and methods to facilitate submission of user images descriptive of locations | |
EP3396566A1 (en) | Method, information processing apparatus and program | |
CN107766881B (en) | Way finding method and device based on basic classifier and storage device | |
CN103399900A (en) | Image recommending method based on location service | |
CN104615620B (en) | Map search kind identification method and device, map search method and system | |
CN108549649B (en) | Rural tourism recommendation method and system based on seasonal characteristics and position characteristics | |
US11127199B2 (en) | Scene model construction system and scene model constructing method | |
WO2023226819A1 (en) | Data matching method and apparatus, readable medium, and electronic device | |
Phan et al. | Collaborative recommendation of photo-taking geolocations | |
Zhang et al. | Wild plant data collection system based on distributed location | |
US11651280B2 (en) | Recording medium, information processing system, and information processing method | |
CN107038169B (en) | Object recommendation method and object recommendation device |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |