CN108062366A - Public culture information recommendation system - Google Patents
Public culture information recommendation system Download PDFInfo
- Publication number
- CN108062366A CN108062366A CN201711287156.2A CN201711287156A CN108062366A CN 108062366 A CN108062366 A CN 108062366A CN 201711287156 A CN201711287156 A CN 201711287156A CN 108062366 A CN108062366 A CN 108062366A
- Authority
- CN
- China
- Prior art keywords
- cultural
- activity
- culture
- venue
- matching
- 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.)
- Granted
Links
- 230000000694 effects Effects 0.000 claims abstract description 359
- 238000001914 filtration Methods 0.000 claims description 41
- 239000011159 matrix material Substances 0.000 claims description 29
- 230000002123 temporal effect Effects 0.000 claims description 23
- 238000000034 method Methods 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 20
- 230000008859 change Effects 0.000 claims description 16
- 238000010276 construction Methods 0.000 claims description 8
- 238000002372 labelling Methods 0.000 claims description 8
- 230000015572 biosynthetic process Effects 0.000 claims description 4
- 238000007619 statistical method Methods 0.000 claims description 4
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 4
- 230000006399 behavior Effects 0.000 description 19
- 230000009471 action Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000003542 behavioural effect Effects 0.000 description 2
- 238000010224 classification analysis Methods 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 241001124569 Lycaenidae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000013549 information retrieval technique Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
- G06F16/9562—Bookmark management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to field of computer technology, specifically provide a kind of Public Culture information recommendation system, it is intended to solve how to participate in user according to the current contextual information of user the technical issues of cultural activity is made overall planning with the behavior of visit culture venue.For this purpose, the interest coalignment of Public Culture information recommendation system can obtain the matching culture keyword, matching culture type label and matching cultural activity form label of active user in the present invention；Cultural venue geography information computing device can position venue, path planning and calculate the distance time；Cultural activity recommendation apparatus can recommended candidate cultural activity；Cultural venue recommendation apparatus can recommend goal culture venue and goal culture activity；Stroke recommendation apparatus can be by goal culture venue, goal culture activity, traffic programme information and time planning information recommendation to user.Cultural activity, cultural venue, traffic, time can be made overall planning by the present invention and recommend user.
Description
Technical field
The present invention relates to field of computer technology, and in particular to a kind of Public Culture information recommendation system.
Background technology
Commending system is click behavior when webpage is browsed according to user and the Characteristic of Interest to internet resource, to user
Recommend its interested relevant information.Commending system is a complicated system engineering, relies on user data, framework, algorithm, man-machine
The combination of the links such as interaction is, it is necessary to the unified knot of the subjects such as statistical technique, data mining technology, information retrieval technique
It closes.
With continuous pursuit of the raising and people of people's economic level to nourishment for the mind, flourishing hair is presented in cultural activity
The situation of exhibition.The type of cultural activity is more and more, such as music, the fine arts, literature, drama, dancing, custom, history and philosophy
Deng at the same time, the form of cultural activity is also varied, such as culture exhibition, cultural performance, cultural exchanges, culture lecture, text
Change publicity, cultural race etc..On the other hand, people are not only only involved in local cultural activity, and more and more people's selection is various
The vehicles trend more great scope of land of various kinds goes to participate in oneself interested cultural activity, and it is interested to make a visit to oneself
Cultural venue.
But with cultural activity type and the blowout of cultural activity form growth and people's pace of life it is fast
Rhythmization, more and more people need a kind of contextual information current according to user to cultural activity, cultural activity venue, traffic
The Public Culture information recommendation system made overall planning and recommended with the time.
The content of the invention
In order to solve the above problem of the prior art, be solve how according to the current contextual information of user to
The technical issues of behavior that family participates in the cultural venue of cultural activity and visit is made overall planning and recommended, the present invention provides one
Kind Public Culture information recommendation system.
Public Culture information recommendation system in the present invention, including：Interest coalignment, cultural venue geography information calculate
Device, cultural activity recommendation apparatus, cultural venue recommendation apparatus and stroke recommendation apparatus；
The interest coalignment is configured to according to the interested cultural vocabulary in user behavior data storehouse and active user,
And the matching culture keyword, matching culture type label and matching cultural activity of active user are obtained using Word2Vec models
Form label；
The culture venue geography information computing device, is configured to according to current geographic position data and cultural venue information
Storehouse carries out the cultural venue in the cultural venue information storehouse venue positioning, planning current geographic position to the cultural field
The path in shop and according to current geographic position described in the path computing planned to the distance time of the culture venue；
The cultural activity recommendation apparatus, be configured to according to active user set time interval, cultural activity information bank,
The current use that current time data, current weather data, the cultural venue information storehouse and the interest coalignment obtain
Matching culture keyword, matching culture type label and the matching cultural activity form label at family, recommended candidate cultural activity；
The culture venue recommendation apparatus is configured to according to the current time data, the cultural venue information storehouse, institute
State the distance time of cultural venue geography information computing module acquisition and candidate's text of cultural activity recommending module acquisition
Goal culture venue and goal culture activity are recommended in change activity；
The stroke recommendation apparatus is configured to the venue position obtained according to the cultural venue geography information computing device
It puts, the goal culture venue that path and distance time and the cultural venue recommendation apparatus obtain and goal culture activity, really
Determine traffic programme information and time planning information and recommend active user.
Preferably, the interest coalignment includes interest labeling module and interest matching module；
The interest labeling module is configured to according to the user behavior data storehouse and default label information, and utilizes system
Meter method carries out statistical analysis, obtains interest tags storehouse；
The interest matching module is configured to according to the interested cultural vocabulary of active user and utilizes Word2Vec models
The interest tags storehouse is matched, obtains the matching culture keyword, matching culture type label and matching culture of active user
Activity form label.
Preferably, the interest matching module includes building of corpus unit, model training unit, term vector acquisition list
Member, similarity calculated and matching unit；
The building of corpus unit is configured in being carried out using the cultural activity information in the cultural activity information bank
Text participle, and corpus is built after carrying out data cleansing according to default deactivated vocabulary, wherein, the default deactivated vocabulary includes one
A or multiple default stop words；
The model training unit is configured to the corpus according to the building of corpus cell formation, and using random
Gradient descent algorithm is trained the Word2Vec models until the Word2Vec models are restrained；
The term vector acquiring unit is configured to using the Word2Vec models after the model training module training,
Obtain the corresponding term vector collection of corpus of the building of corpus cell formation；
The similarity calculated, is configured so that Pearson correlation coefficient, calculates the term vector acquiring unit and obtains
The term vector collection taken, respectively with cultural keyword, culture type label and the cultural activity form mark in the interest tags storehouse
The similarity of label；
It is crucial more than the culture of threshold value to be configured to the similarity for obtaining the similarity calculated for the matching unit
Word, culture type label and cultural activity form label respectively as the cultural keyword of matching, matching culture type label and
With cultural activity form label.
Preferably, the cultural activity recommendation apparatus includes cultural activity semantic modules and cultural activity recommending module；
The cultural activity semantic modules are configured to obtain the culture pass of the cultural activity in the cultural activity information bank
Keyword, culture type label and cultural activity form label；
The cultural activity recommending module is configured to the time interval, the cultural activity letter that are set according to the active user
The matching culture for the active user that breath storehouse, current time data, current weather data, the interest coalignment obtain is crucial
Word, matching culture type label and the culture for matching cultural activity form label and cultural activity semantic modules acquisition
Keyword, culture type label and cultural activity form label, obtain candidate's cultural activity.
Preferably, the cultural activity recommending module includes cultural activity temporal filtering unit, cultural activity weather filters
Unit and matching value computing unit；
The cultural activity temporal filtering unit is configured to the time interval set according to the active user and the text
Change action message storehouse, obtain the first cultural activity in the time interval set in the active user；
The cultural activity weather filter element is configured to the first text for obtaining the cultural activity temporal filtering unit
Change activity, according to the current weather data, cultural activity information bank and the cultural venue information storehouse, acquisition meets current
The second culture activity of weather condition；
The matching value computing unit is configured to according to method shown in following formula, calculates the second culture activity and institute
State the matching value between the cultural keyword of matching, matching culture type and the matching cultural activity form of interest coalignment acquisition
D, and descending sort is carried out according to the matching value, the 1-M cultural activity is chosen as candidate's cultural activity, wherein M >=1：
Wherein, α is the weight for matching cultural activity form, and β is the weight for matching culture type, and γ is that matching culture is crucial
The weight of word, α, beta, gamma ∈ [0,1], and alpha+beta+γ=1；
N is the quantity of the cultural activity form label of the second culture activity, and m is that the interest coalignment obtains
Match the quantity of cultural activity form label；Match(ATi,ATl') for i-th of cultural activity in the second culture activity
Form label A TiCultural activity form label A T is matched with described l-thl' Boolean, and as the ATiWith the ATl' one
During cause, Match (ATi,ATl')=1；When inconsistent, Match (ATi,ATl')=0；；
K is the quantity of the culture type label of the second culture activity, and g is the matching that the interest matching module obtains
The quantity of culture type label；Match(CTu,CTv') for u-th of culture type label C T in the second culture activityuWith
L-th of matching culture type label C Tv' Boolean, and as the CTuWith the CTv' it is consistent when, Match (CTu,
CTv')=1；When inconsistent, Match (CTu,CTv')=0；
P is the quantity of the cultural keyword of the second culture activity, and q is the matching text that the interest coalignment obtains
Change the quantity of keyword；KWxFor x-th of cultural keyword in the second culture activity, KW 'yIt is closed for y-th of matching culture
Keyword；
Sim is similarity function, and Sim ∈ (0,1], M is candidate's cultural activity recommended amount of the system default.
Preferably, the cultural venue recommendation apparatus includes venue matching module, venue temporal filtering module and venue road
Way temporal filtering module；
The venue matching module is configured to the candidate's cultural activity obtained according to the cultural activity recommendation apparatus and institute
Cultural venue information storehouse is stated, matches the cultural venue of candidate's cultural activity；
The venue temporal filtering module, be configured to according to the current time data, the cultural venue information storehouse with
The cultural venue for candidate's cultural activity that the venue matching module obtains is filtered, and obtains the first cultural venue；
The venue travel time filtering module is configured to according to the cultural venue information storehouse with the cultural venue
The distance time that information computing device obtains is managed, the first cultural venue that the venue temporal filtering module obtains was carried out
Filter obtains goal culture venue and goal culture activity.
Preferably, the cultural venue recommendation apparatus further includes standing cultural activity recommending module；The standing culture is living
Dynamic recommending module is configured to obtain using collaborative filtering similar with current user interest in the user behavior data storehouse
User group, and recommend the interested cultural activity of user group described in the target venue to active user.
Preferably, it is single to include user group acquiring unit, User Activity matrix structure for the standing cultural activity recommending module
Member, interest value computing unit and standing cultural activity recommendation unit；
The user group acquiring unit is configured to the goal culture field obtained according to the venue travel time filtering module
Shop and the user behavior data storehouse obtain the user group for once accessing the goal culture venue；
The User Activity matrix construction unit is configured to according to the user group that the user group acquiring unit obtains with working as
The cultural activity historical data of preceding user builds user-active matrix；
The interest value computing unit is configured to using LFM matrix decomposition methods, by the User Activity matrix construction unit
User-active matrix of acquisition calculates user's matrix and cultural activity matrix, and according to user's matrix and the culture
Active matrix calculates interest value of the active user to each cultural activity；
The standing cultural activity recommendation unit is configured to the active user couple obtained according to the interest value computing unit
Cultural activity in the goal culture venue is carried out descending sort, and chooses 1-N by the interest value of each cultural activity
Cultural activity in goal culture venue is recommended, and wherein N is the standing cultural activity recommended amount of the system default, and
N≥1。
Preferably, the stroke recommendation apparatus includes traffic programme module and time planning module；
The traffic programme module is configured to the goal culture venue obtained according to the cultural venue recommendation apparatus and profit
With the cultural venue geography information computing device, the geo-location and path planning of venue are carried out, obtains traffic programme information；
The time planning module is configured to perform operations described below：
Step A1, the goal culture venue obtained according to the cultural venue recommendation apparatus, goal culture activity with it is described
The traffic programme information that traffic programme module obtains, and venue distance is calculated by the cultural venue geography information computing device
Time；
Step A2, the time interval set according to the active user, venue distance time and shown according to the following formula
Method obtains the total time that active user participates in goal culture activity：
t3=t1-t2
Wherein, t1For the time interval that the active user is set, t2For the venue distance time, t3For active user
Participate in the total time of goal culture activity；
Step A3, according to the active user that the standing cultural activity recommendation unit obtains to the interest of each cultural activity
Value calculates each goal culture activity and accounts for the time scale that the active user participates in the total time of goal culture activity；
Step A4 according to the time scale, the cultural venue information storehouse and the cultural activity information bank, plans institute
The participation time of each goal culture activity is stated, obtains time planning information.
Preferably, the Public Culture information recommendation system further includes feedback device；The feedback device is configured to work as institute
It states active user and receives the information that the stroke recommendation apparatus is recommended, the information that the stroke recommendation apparatus is recommended is charged to described
User behavior data storehouse.
Compared with the immediate prior art, above-mentioned technical proposal at least has the advantages that：
1. in the Public Culture information recommendation system of the present invention, by setting interest coalignment, culture within the system
Venue geography information computing device, cultural activity recommendation apparatus, cultural venue recommendation apparatus and stroke recommendation apparatus, can basis
User's current time data, geodata and weather condition, by cultural venue recommendation information, cultural activity recommendation information, when
Between planning information and traffic programme information made overall planning and recommend user, user is allowed to save worry, and to participate in culture living for labour-saving
The cultural venue of dynamic and visit.
It, can be with using the cultural activity recommendation apparatus in the system 2. in the Public Culture information recommendation system of the present invention
Cultural activity in the time interval and cultural activity information bank that are set according to active user is held temporal filtering and is fallen not current
Cultural activity in the time interval of user setting；The venue held by cultural activity information bank chinesization activity works as the day before yesterday
Venue in destiny evidence and cultural venue storehouse is indoor or outdoor information filtering is fallen not meeting the cultural activity of weather condition；
Matching culture keyword, matching culture type and the matching that remaining cultural activity after filtering and interest coalignment are obtained
Activity form is matched, and finally draws candidate's cultural activity recommended to the user, so that the system recommends to meet to user
User time meets current weather condition and is the interested cultural activity of user.
3. in the Public Culture information recommendation system of the present invention, using the cultural venue recommendation apparatus in the system, according to
The field that venue essential information, venue open hour in cultural venue information storehouse are obtained with cultural venue geography information computing device
The shop travel time, the candidate's cultural activity obtained to cultural activity recommendation apparatus is filtered, during so as to exclude not meeting user
Between cultural activity and cultural venue, when obtaining meeting the time of active user, meeting the venue open hour, meet venue road
Between and be the interested cultural activity of user and cultural venue, further improve user and participate in cultural activity and visit culture
The user experience of venue.
4. in the Public Culture information recommendation system of the present invention, using the standing cultural activity recommending module in the system,
The user group similar with current user interest is calculated, and then show that the user group is interested standing in goal culture venue
Cultural activity simultaneously recommends active user so that it is convenient to which user participates in the interested culture work of its possibility in same cultural venue
It is dynamic, it is final that active user is allowd to participate in more interested cultural activities within the limited time.
5. in the Public Culture information recommendation system of the present invention, by setting feedback device within the system, when user connects
By the system recommendation information when, this recommendation information is charged into user behavior data storehouse, so as to allow number in user behavior data
According to more precisely reliable.
Description of the drawings
Fig. 1 is the Public Culture information recommendation system major architectural schematic diagram of the embodiment of the present invention；
In attached drawing mark for：100- interest coalignments, 110- interest labeling modules, 120- interest matching modules, 121-
Building of corpus unit, 122- model training units, 123- term vector acquiring units, 124- similarity calculateds, 125-
With unit, 200- culture venue geography information computing devices, 300- cultural activity recommendation apparatus, 310- cultural activity semanteme moulds
Block, 311- keyword abstraction units, 312- culture types mark unit, 313- activity forms mark unit, 320- cultural activities
Recommending module, 321- cultural activity temporal filtering units, 322- cultural activity weather filter elements, 323- matching values calculate single
Member, 400- culture venue recommendation apparatus, 410- venue matching modules, 420- venue temporal filtering modules, during 430- venue roads
Between filtering module, 440- sets up cultural activity recommending module, 441- user group acquiring units, and 442- User Activities matrix structure is single
Member, 443- interest value computing units, 444- set up cultural activity recommendation unit, 500- stroke recommendation apparatus, 510- traffic programmes
Module, 520 time planning modules, 600- context data harvesters, 700- input units, 800- feedback devices.
Specific embodiment
The preferred embodiment of the present invention described with reference to the accompanying drawings.It will be apparent to a skilled person that this
A little embodiments are used only for explaining the technical principle of the present invention, it is not intended that limit the scope of the invention.
Public Culture information recommendation system in the present invention is according to the current context data of user, such as current time number
According to, current geographic position data and current weather data, and according to user behavior data storehouse, cultural activity information bank and cultural field
Shop information bank recommends the commending system of cultural activity, cultural venue, traffic programme information and time planning information to user, wherein
The culture information recommendation system is suitable for the information recommendation of time within one day.
Below in conjunction with the accompanying drawings, Public Culture information recommendation system in the embodiment of the present invention is illustrated.
Refering to attached drawing 1, Fig. 1 illustratively shows the main frame of Public Culture information recommendation system of the embodiment of the present invention
Frame.As shown in Figure 1, Public Culture information recommendation system be with can including interest coalignment 100, cultural venue in the present embodiment
Manage information computing device 200, cultural activity recommendation apparatus 300, cultural venue recommendation apparatus 400 and stroke recommendation apparatus 500.
Further, interest coalignment 100 is configurable to according to user behavior data storehouse and current in the present embodiment
The interested cultural vocabulary of user, and utilize the matching culture keyword of Word2Vec models acquisition active user, matching culture
Type label and matching cultural activity form label.
Further, interest coalignment 100 can include interest labeling module 110 and interest matching mould in the present embodiment
Block 120.
Wherein, interest labeling module 110 is configurable to according to user behavior data storehouse and default label information, and is utilized
Statistical method carries out statistical analysis, obtains interest tags storehouse.
Specifically, user behavior data storehouse includes the behavioral data of user's participation cultural activity and user's ginseng in the present embodiment
See the behavioral data of cultural venue.The interest tags storehouse that interest labeling module 110 obtains includes cultural keyword label collection, culture
Type label collection and cultural activity form tally set.Cultural keyword, culture type label and cultural activity form label are all
Textual form, but the scope of cultural keyword is not fixed, culture type label is the culture according to default definition for tag information
Type, cultural activity form label are the cultural activity forms according to default definition for tag information.
Interest matching module 120 is configurable to according to the interested cultural vocabulary of active user and utilizes Word2Vec moulds
Type match interest tag library, the matching culture keyword, matching culture type label and matching culture for obtaining active user are living
Dynamic form label.
Specifically, the matching culture keyword in the present embodiment is the interested cultural vocabulary of active user and interest tags
It is after cultural keyword set matching in storehouse as a result, matching culture type is the interested cultural vocabulary of active user and interest mark
It signs after the culture type tally set matching in storehouse as a result, matching cultural activity form label is the interested culture of active user
The result after the matching of cultural activity form tally set in vocabulary and interest tags storehouse.
Further, interest matching module 120 includes building of corpus unit 121, model training unit in this implementation
122nd, term vector acquiring unit 123, similarity calculated 124 and matching unit 125.
Wherein, building of corpus unit 121 be configurable to using the cultural activity information in cultural activity information bank into
Row Chinese word segmentation, and corpus is built after carrying out data cleansing according to default deactivated vocabulary, wherein, presetting deactivated vocabulary includes one
A or multiple default stop words.
Specifically, the present embodiment chinesization action message library storage the information of the current cultural activity held, tool
The title of body including cultural activity, cultural activity hold the time, cultural activity hold venue, the word of cultural activity is situated between
It continues, cultural keyword, culture type and cultural activity form, wherein culture type can be music, the fine arts, literature, drama, dance
It steps, custom, history, philosophy etc.；Cultural activity form can be culture exhibition, cultural performance, cultural exchanges, cultural lecture, text
Change publicity, cultural race etc..All comprising information above, each cultural activity has for each cultural activity in cultural activity information bank
One group of culture keyword, each cultural activity can there are one or multiple culture types, each cultural activity can there are one or
Multiple cultural activity forms.The identical cultural activity of content to hold the time different or hold venue difference, be considered as different
Cultural activity.
Default deactivated vocabulary is Manual definition, for filtering out the word in default deactivated vocabulary.According to default stop words
Table is filtered every words or phrase, keeps original order constant the effective vocabulary obtained after filtering, is used between vocabulary
Space is isolated, and obtains orderly phrase.Phrase is joined end to end afterwards, is isolated between phrase using space,
The big phrase of an entirety is formed, is corpus.
Model training unit 122 is configurable to the corpus according to 121 structure of building of corpus unit, and using random
Gradient descent algorithm is trained Word2Vec models until Word2Vec models are restrained.
Specifically, Word2Vec models are the Natural Language Processing Models based on deep learning in the present embodiment.
Term vector acquiring unit 123 is configurable to use the Word2Vec models after the training of model training unit 122,
Obtain the corresponding term vector collection of corpus of 121 structure of building of corpus unit.
Similarity calculated 124 is configurable to using Pearson correlation coefficient, is calculated term vector acquiring unit 123 and is obtained
The term vector collection taken, respectively with cultural keyword, culture type label and the cultural activity form label in interest tags storehouse
Similarity.
Matching unit 125 is configurable to culture pass of the similarity for obtaining similarity calculated 124 more than threshold value
Keyword, culture type label and cultural activity form label respectively as the cultural keyword of matching, matching culture type label and
Match cultural activity form label.
Specifically, cultural keyword is matched in this implementation and includes one or more cultural keywords, matches culture type mark
Label include one or more culture type labels, and matching cultural activity form label includes one or more cultural activity form marks
Label.
Cultural venue geography information computing device 200 is configured to according to current geographic position data and cultural venue information
Storehouse carries out the cultural venue in cultural venue information storehouse venue positioning, planning current geographic position to the path of cultural venue
And according to the path computing current geographic position planned to the distance time of cultural venue.
Specifically, current geographic position data can be by being equipped with the terminal device or geography information of sensor in this implementation
System obtains；Terminal device can be mobile phone, tablet computer, laptop etc., and GIS-Geographic Information System can be Baidu map
API.Wherein, the distance time of current geographic position to cultural venue can be according to the vehicles and the culture field that user takes
What the path that shop geography information computing device 200 is planned obtained.
Cultural venue information library storage cultural venue information, the essential information of venue is specifically included, such as venue's name, field
Shop indoors or outdoor etc., the cultural activity that venue open hour, venue geographical location and venue are set up, such as standing exhibition.
Cultural venue geography information computing device 200 in the present embodiment is based on the venue geography position in cultural venue information storehouse
It puts.
Cultural activity recommendation apparatus 300, be configured to according to active user set time interval, cultural activity information bank,
The active user's that current time data, current weather data, cultural venue information storehouse and interest coalignment 100 obtain
The cultural keyword of matching, matching culture type label and matching cultural activity form label, recommended candidate cultural activity.
Specifically, current time data, current weather data can be by networks or equipped with sensor in the present embodiment
Terminal device obtain, wherein terminal device can be mobile phone, tablet computer, laptop etc..
Further, the present embodiment chinesization activity recommendation device 300 includes cultural activity semantic modules 310 and culture is lived
Dynamic recommending module 320.
Wherein, cultural activity semantic modules 310 are configurable to obtain the text of the cultural activity in cultural activity information bank
Change keyword, culture type label and cultural activity form label.
Specifically, this implementation chinesization Activity semantics module 310 includes keyword abstraction unit 311, culture type marks
Unit 312 and activity form mark unit 313.
Wherein, keyword abstraction unit 311 is configurable to carry out cultural activity information bank using TextRank algorithm
The keyword of cultural activity is extracted in text analyzing.
Specifically, the keyword abstraction unit 311 in the present embodiment is to the cultural activity in cultural activity information bank
Title and the character introduction of cultural activity carry out text analyzing.
Culture type mark unit 312 is configurable to through the first disaggregated model to the culture in cultural activity information bank
Activity carries out classification analysis, obtains culture type label.
Specifically, the first disaggregated model is Bayesian Classification Model in this implementation, and text is obtained using the first disaggregated model
It also needs to be trained the first disaggregated model before changing type label.The optimization training step of first disaggregated model includes step
S11, step S12 and step S13.
By way of manually marking, culture type mark is marked to the cultural activity in cultural activity information bank by step S11
Label, obtain culture type training set and culture type test set.
Step S12 is sampled from culture type training set, is classified using the culture type training set after sampling to first
Model is trained.
Step S13 carries out assessment until first point using culture type test set to the training result of the first disaggregated model
The accuracy rate and recall rate of the output result of class model reach preset first threshold value, otherwise adjust the to be optimized of the first disaggregated model
The weighted value of parameter is simultaneously again trained it.
Activity form mark unit 313 is configurable to through the second disaggregated model to the culture in cultural activity information bank
Activity carries out classification analysis, obtains cultural activity form label.
Specifically, the second disaggregated model is Bayesian Classification Model in this implementation, and text is obtained using the second disaggregated model
It also needs to be trained the second disaggregated model before changing activity form label.The optimization training step of second disaggregated model includes step
Rapid S21, step S22 and step S23.
By way of manually marking, cultural activity shape is marked to the cultural activity in cultural activity information bank by step S21
Formula label obtains cultural activity formal disciform collection and cultural activity formal testing collection.
Step S22 is sampled from cultural activity formal disciform collection, utilizes the cultural activity formal disciform set pair after sampling
Second disaggregated model is trained.
Step S23 carries out assessment until the using the training result of cultural activity formal testing the second disaggregated model of set pair
The accuracy rate and recall rate of the output result of two disaggregated models reach default second threshold, otherwise adjust treating for the second disaggregated model
The weighted value of Optimal Parameters is simultaneously again trained it.
Cultural activity recommending module 320, be configured to according to active user set time interval, cultural activity information bank,
The matching culture keyword for the active user that current time data, current weather data, interest coalignment 100 obtain, matching
Cultural keyword, the text that culture type label and matching cultural activity form label and cultural activity semantic modules 310 obtain
Change type label and cultural activity form label, obtain candidate's cultural activity.
Further, the cultural activity recommending module 320 in this implementation includes cultural activity temporal filtering unit 321, text
Change activity weather filter element 322 and matching value computing unit 323.
Wherein, cultural activity temporal filtering unit 321 is configurable to the time interval and text that are set according to active user
Change action message storehouse, obtain the first cultural activity in the time interval set in active user.
Specifically, the time interval that active user is set in this implementation is that user inputs working as the culture information recommendation system
The effective time of preceding user.Cultural activity temporal filtering unit 321 can be according to the time interval and culture that active user is set
The time is held in cultural activity in action message storehouse, is filtered, so as to obtain in the time interval set in active user
The first cultural activity.
Cultural activity weather filter element 322 is configurable to obtain cultural activity temporal filtering unit 321 first
Cultural activity meets current weather feelings according to current weather data, cultural activity information bank and cultural venue information storehouse, acquisition
The second culture activity of condition.
Specifically, the cultural activity weather filter element 322 in this implementation can be living according to current weather data, culture
The venue held in venue and cultural venue information storehouse of cultural activity in dynamic information bank is indoors or outdoor information
It is filtered, so as to obtain the second culture activity for meeting current weather conditions.Wherein, current weather data can be by working as
The terminal device of preceding user obtains, and can also be obtained by network.
Matching value computing unit 323 is configurable to according to the method shown in following formula (1), calculate second culture activity with it is emerging
Matching value between the cultural keyword of matching, matching culture type and matching cultural activity form that interesting coalignment 100 obtains
D, and descending sort is carried out according to matching value, the 1-M cultural activity is chosen as candidate's cultural activity, wherein M >=1：
Wherein, α is the weight for matching cultural activity form, and β is the weight for matching culture type, and γ is that matching culture is crucial
The weight of word, α, beta, gamma ∈ [0,1], and alpha+beta+γ=1；
N is the quantity of the cultural activity form label in second culture activity, and m is the matching text that interest coalignment obtains
Change the quantity of activity form label；I is the index of the cultural activity form label of second culture activity, and l is interest coalignment
The index of the matching cultural activity form label of acquisition；Match(ATi,ATl') live for i-th of culture in second culture activity
Dynamic form label A TiCultural activity form label A T is matched with l-thl' Boolean, and work as ATiWith ATl' it is consistent when, Match
(ATi,ATl')=1；When inconsistent, Match (ATi,ATl')=0；
K is the quantity of the culture type label of second culture activity, and g is the matching culture type that interest matching module obtains
The quantity of label；U is the index of the culture type label of second culture activity, and v is the matching culture that interest coalignment obtains
The index of type label；Match(CTu,CTv') for u-th of culture type label C T in second culture activityuWith l-th
With culture type label C Tv' Boolean, and work as CTuWith CTv' it is consistent when, Match (CTu,CTv')=1；When inconsistent,
Match(CTu,CTv')=0；
P is the quantity of the cultural keyword of second culture activity, and q is the matching culture keyword that interest coalignment obtains
Quantity；X is the cultural keyword index of second culture activity, and y is the cultural keyword of matching that interest coalignment obtains
Index；KWxFor in second culture activity
X-th of cultural keyword, KW 'yFor the cultural keyword of y-th of matching；
Sim is similarity function, and Sim ∈ (0,1], M is candidate's cultural activity recommended amount of system default.
Specifically, α in the present embodiment, the parameter value of beta, gamma vary with each individual, α, and the parameter value of beta, gamma is according to each user
Participate in the optimal value that the historical data of cultural activity is trained by least square method.When data volume is smaller in the early stage
It waits, α, three parameters of beta, gamma are disposed as 0.333.
Calculating the method for the crucial Word similarity of culture includes step S31 and step S32.
Step S31 obtains the term vector of the cultural keyword of second culture activity and matching text by Word2Vec models
Change the term vector of keyword.
Step S32 calculates term vector and the matching of the cultural keyword of second culture activity by cosine similarity method
The similarity of the term vector of cultural keyword obtains the matching degree between cultural keyword.
Cultural venue recommendation apparatus 400 is configured to according to current time data, cultural venue information storehouse, cultural venue
The distance time of information computational module acquisition and candidate's cultural activity of the acquisition of cultural activity recommending module 320 are managed, recommends mesh
The cultural venue of mark and goal culture activity.
Further, the cultural venue recommendation apparatus 400 in the present embodiment includes venue matching module 410, venue time
Filtering module 420 and venue travel time filtering module 430.
Wherein, candidate's culture that venue matching module 410 is configurable to obtain according to cultural activity recommendation apparatus 300 is lived
Dynamic and cultural venue information storehouse, the cultural venue of matching candidate cultural activity.
Specifically, the venue matching module 410 in the present embodiment is according to candidate's cultural activity and cultural venue information storehouse
In venue essential information, if venue's name, venue are indoors or outdoor information carries out the matched of cultural venue.
Venue temporal filtering module 420 is configurable to according to current time data, cultural venue information storehouse and venue
The cultural venue of candidate's cultural activity with the acquisition of module 410 is filtered, and obtains the first cultural venue.
Specifically, the venue temporal filtering module 420 in the present embodiment is according to current time data, cultural venue information
Venue open hour and candidate's cultural activity corresponding cultural venue in storehouse is filtered, and obtains the first cultural venue.
Venue travel time filtering module 430 is configurable to according to cultural venue information storehouse and cultural venue geography information
The distance time that computing device 200 obtains is filtered the first cultural venue that venue temporal filtering module 420 obtains, obtains
Obtain goal culture venue and goal culture activity.
Specifically, the venue travel time filtering module 430 in the present embodiment is in the cultural venue information storehouse of foundation
The distance time that venue open hour, cultural venue geography information computing device 200 obtain is filtered the first cultural venue,
So as to obtain goal culture venue and goal culture activity.
Further, the cultural venue recommendation apparatus 400 in the present embodiment can also include standing cultural activity recommendation mould
Block 440；Standing cultural activity recommending module 440 be configurable to using collaborative filtering obtain in user behavior data storehouse with
The similar user group of current user interest, and recommend in target venue the interested cultural activity of user group to active user.
Further, the standing cultural activity recommending module 440 in the present embodiment includes user group acquiring unit 441, uses
Family active matrix construction unit 442, interest value computing unit 443 and standing cultural activity recommendation unit 444.
Wherein, user group acquiring unit 441 is configurable to the target obtained according to venue travel time filtering module 430
Cultural venue and user behavior data storehouse obtain the user group for once accessing goal culture venue.
User Activity matrix construction unit 442 be configurable to according to the user group that user group acquiring unit 441 obtains with
The cultural activity historical data of active user builds user-active matrix.
Interest value computing unit 443 is configurable to using LFM matrix decomposition methods, by User Activity matrix construction unit
442 user-the active matrixs obtained calculate user's matrix and cultural activity matrix, and according to user's matrix and cultural activity square
Battle array calculates interest value of the active user to each cultural activity.
Standing cultural activity recommendation unit 444 is configurable to the active user obtained according to interest value computing unit 443
To the interest value of each cultural activity, the cultural activity in goal culture venue is subjected to descending sort, and chooses the 1-N mesh
Cultural activity in the cultural venue of mark is recommended, and wherein N is the standing cultural activity recommended amount of system default, and N >=1.
Stroke recommendation apparatus 500 is configurable to the venue position obtained according to cultural venue geography information computing device 200
It puts, the goal culture venue that path and distance time and cultural venue recommendation apparatus 400 obtain and goal culture activity, really
Determine traffic programme information and time planning information and recommend active user.
Further, the stroke recommendation apparatus 500 in the present embodiment can include traffic programme module 510 and time planning
Module 520.
Wherein, traffic programme module 510 is configurable to the goal culture field obtained according to cultural venue recommendation apparatus 400
Shop simultaneously utilizes cultural venue geography information computing device 200, carries out the geo-location and path planning of venue, obtains traffic programme
Information.
Time planning module 520 is configurable to perform operations described below：
Step A1, the goal culture venue obtained according to cultural venue recommendation apparatus 400, goal culture activity are advised with traffic
When drawing the traffic programme information that module 510 obtains, and venue distance calculated by cultural venue geography information computing device 200
Between.
Step A2, according to the time interval of active user's setting, venue distance time and method shown in (2) obtains according to the following formula
Active user is taken to participate in the total time of goal culture activity：
t3=t1-t2 (2)
Wherein, t1For the time interval that active user is set, t2For venue distance time, t3Target is participated in for active user
The total time of cultural activity.
Step A3, according to the active user that standing cultural activity recommendation unit 444 obtains to the interest of each cultural activity
Value calculates each goal culture activity and accounts for the time scale that active user participates in the total time of goal culture activity.
Step A4 according to time scale, cultural venue information storehouse and cultural activity information bank, plans that each goal culture is lived
The dynamic participation time obtains time planning information.
Further, the Public Culture information recommendation system in the present embodiment can also include context data harvester
600；Context data harvester 600 be configurable to by terminal obtain current time data, current geographic position data with
Current weather data.
Further, the Public Culture information recommendation system in the present embodiment can also include input unit 700；Input dress
It puts 700 and is configurable to time interval and the interested cultural vocabulary of active user that acquisition active user is set.
Further, the Public Culture information recommendation system in the present embodiment can also include feedback device 800；Feedback dress
It puts 800 to be configurable to receive the information of the recommendation of stroke recommendation apparatus 500 as active user, stroke recommendation apparatus 500 is recommended
Information charge to user behavior data storehouse.
Specifically, the feedback device 800 in the present embodiment can make the information in user's behavior database more perfect, separately
On the one hand because the data of the record in user behavior data storehouse are because obtained the accreditation of user, user behavior data storehouse
In data can be more accurate.
It will be understood by those skilled in the art that above-mentioned Public Culture information recommendation system further includes some other known knots
Structure, such as processor, controller, memory etc., wherein, memory includes but not limited to random access memory, flash memory, read-only storage
Device, programmable read only memory, volatile memory, nonvolatile memory, serial storage, parallel storage or register
Deng, processor includes but not limited to CPLD/FPGA, DSP, arm processor, MIPS processors etc., in order to unnecessarily obscure this
Disclosed embodiment, these well known structures are not shown.
It will be understood by those skilled in the art that the module in the device in embodiment can adaptively be changed
And they are arranged in one or more devices different from the embodiment.Can the module in embodiment or unit or
Component is combined into a module or unit or component and can be divided into multiple submodule or subelement or subgroup in addition
Part.In addition at least some in such feature and/or process or unit exclude each other, any combinations may be employed
To all features disclosed in this specification (including adjoint claim, summary and attached drawing) and such disclosed any side
All processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (will including adjoint right
Ask, make a summary and attached drawing) disclosed in each feature can be replaced by the alternative features for providing identical, equivalent or similar purpose.
The all parts embodiment of the present invention can be with hardware realization or to be run on one or more processor
Software module realize or realized with combination thereof.It will be understood by those of skill in the art that it can use in practice
Microprocessor or digital signal processor (DSP) realize some in server according to embodiments of the present invention, client
Or some or all functions of whole components.The present invention is also implemented as performing the one of method as described herein
Partly or completely equipment or program of device (for example, PC programs and PC program products).Such journey for realizing the present invention
Sequence can be stored on PC readable mediums or can have the form of one or more signal.Such signal can be from
It is downloaded on internet website and obtains either providing on carrier signal or providing in the form of any other.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included some features rather than other feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in claims of the present invention, embodiment claimed
It is one of arbitrary mode to use in any combination.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.The present invention can be realized by means of including the hardware of several different elements and by means of properly programmed PC.
If in the unit claim for listing equipment for drying, several in these devices can be come specific by same hardware branch
It embodies.The use of word first, second, and third does not indicate that any order.These words can be construed to title.
So far, have been combined preferred embodiment shown in the drawings and describe technical scheme, still, this field
Technical staff is it is easily understood that protection scope of the present invention is expressly not limited to these specific embodiments.Without departing from this
On the premise of the principle of invention, those skilled in the art can make correlation technique feature equivalent change or replacement, these
Technical solution after changing or replacing it is fallen within protection scope of the present invention.
Claims (10)
1. a kind of Public Culture information recommendation system, which is characterized in that the system comprises interest coalignment, cultural venues
Manage information computing device, cultural activity recommendation apparatus, cultural venue recommendation apparatus and stroke recommendation apparatus；
The interest coalignment is configured to according to the interested cultural vocabulary in user behavior data storehouse and active user, and profit
The matching culture keyword, matching culture type label and matching cultural activity form of active user is obtained with Word2Vec models
Label；
The culture venue geography information computing device is configured to according to current geographic position data and cultural venue information storehouse,
Venue positioning, planning current geographic position to the cultural venue are carried out to the cultural venue in the cultural venue information storehouse
Path and according to current geographic position described in the path computing planned to the distance time of the culture venue；
The cultural activity recommendation apparatus is configured to the time interval set according to active user, cultural activity information bank, current
The active user's that time data, current weather data, the cultural venue information storehouse and the interest coalignment obtain
The cultural keyword of matching, matching culture type label and matching cultural activity form label, recommended candidate cultural activity；
The culture venue recommendation apparatus is configured to according to the current time data, the cultural venue information storehouse, the text
Change the distance time of venue geography information computing module acquisition and candidate's culture of cultural activity recommending module acquisition is lived
It is dynamic, recommend goal culture venue and goal culture activity；
The stroke recommendation apparatus is configured to the venue location, the road that are obtained according to the cultural venue geography information computing device
The goal culture venue and goal culture activity that footpath and distance time and the cultural venue recommendation apparatus obtain, determine to hand over
Logical planning information and time planning information simultaneously recommend active user.
2. Public Culture information recommendation system according to claim 1, which is characterized in that the interest coalignment includes
Interest labeling module and interest matching module；
The interest labeling module is configured to according to the user behavior data storehouse and default label information, and utilizes statistics side
Method carries out statistical analysis, obtains interest tags storehouse；
The interest matching module is configured to according to the interested cultural vocabulary of active user and utilizes Word2Vec Model Matchings
The interest tags storehouse obtains the matching culture keyword, matching culture type label and matching cultural activity of active user
Form label.
3. Public Culture information recommendation system according to claim 2, which is characterized in that the interest matching module includes
Building of corpus unit, model training unit, term vector acquiring unit, similarity calculated and matching unit；
The building of corpus unit is configured to carry out Chinese point using the cultural activity information in the cultural activity information bank
Word, and corpus is built after carrying out data cleansing according to default deactivated vocabulary, wherein, the default deactivated vocabulary include one or
Multiple default stop words；
The model training unit is configured to the corpus according to the building of corpus cell formation, and using stochastic gradient
Descent algorithm is trained the Word2Vec models until the Word2Vec models are restrained；
The term vector acquiring unit is configured to, using the Word2Vec models after the model training module training, obtain
The corresponding term vector collection of corpus of the building of corpus cell formation；
The similarity calculated, is configured so that Pearson correlation coefficient, calculates what the term vector acquiring unit obtained
Term vector collection, respectively with cultural keyword, culture type label and the cultural activity form label in the interest tags storehouse
Similarity；
The matching unit, be configured to the similarity for obtaining the similarity calculated more than threshold value cultural keyword,
Culture type label and cultural activity form label are respectively as the cultural keyword of matching, matching culture type label and matching text
Change activity form label.
4. Public Culture information recommendation system according to claim 1, which is characterized in that the cultural activity recommendation apparatus
Including cultural activity semantic modules and cultural activity recommending module；
The cultural activity semantic modules are configured to obtain the culture key of the cultural activity in the cultural activity information bank
Word, culture type label and cultural activity form label；
The cultural activity recommending module, be configured to according to the active user set time interval, cultural activity information bank,
The matching culture keyword for the active user that current time data, current weather data, the interest coalignment obtain, matching
Culture type label and matching cultural activity form label and the cultural activity semantic modules obtain cultural keyword,
Culture type label and cultural activity form label, obtain candidate's cultural activity.
5. Public Culture information recommendation system according to claim 4, which is characterized in that the cultural activity recommending module
Including cultural activity temporal filtering unit, cultural activity weather filter element and matching value computing unit；
The cultural activity temporal filtering unit is configured to living according to the time interval of active user setting and the culture
Dynamic information bank, obtains the first cultural activity in the time interval set in the active user；
It is living to be configured to the first culture for obtaining the cultural activity temporal filtering unit for the cultural activity weather filter element
It is dynamic, meet current weather according to the current weather data, cultural activity information bank and the cultural venue information storehouse, acquisition
The second culture activity of situation；
The matching value computing unit is configured to according to method shown in following formula, calculate the second culture activity with it is described emerging
Matching value D between the cultural keyword of matching, matching culture type and matching cultural activity form that interesting coalignment obtains, and
Descending sort is carried out according to the matching value, chooses the 1-M cultural activity as candidate's cultural activity, wherein M >=1：
<mfenced open = "" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>D</mi>
<mo>=</mo>
<mi>&alpha;</mi>
<mo>&CenterDot;</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>l</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>m</mi>
</munderover>
<mi>M</mi>
<mi>a</mi>
<mi>t</mi>
<mi>c</mi>
<mi>h</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>AT</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<msubsup>
<mi>AT</mi>
<mi>l</mi>
<mo>&prime;</mo>
</msubsup>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>+</mo>
<mi>&beta;</mi>
<mo>&CenterDot;</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>u</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>v</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>g</mi>
</munderover>
<mi>M</mi>
<mi>a</mi>
<mi>t</mi>
<mi>c</mi>
<mi>h</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>CT</mi>
<mi>u</mi>
</msub>
<mo>,</mo>
<msubsup>
<mi>CT</mi>
<mi>v</mi>
<mo>&prime;</mo>
</msubsup>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>+</mo>
<mi>&gamma;</mi>
<mo>&CenterDot;</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>x</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>p</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>y</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>q</mi>
</munderover>
<mi>S</mi>
<mi>i</mi>
<mi>m</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>KW</mi>
<mi>x</mi>
</msub>
<mo>,</mo>
<msubsup>
<mi>KW</mi>
<mi>y</mi>
<mo>&prime;</mo>
</msubsup>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
Wherein, α is the weight for matching cultural activity form, and β is the weight for matching culture type, and γ is the cultural keyword of matching
Weight, α, beta, gamma ∈ [0,1], and alpha+beta+γ=1；
N is the quantity of the cultural activity form label of the second culture activity, and m is the matching that the interest coalignment obtains
The quantity of cultural activity form label；Match(ATi,ATl') for i-th of cultural activity form in the second culture activity
Label A TiCultural activity form label A T is matched with described l-thl' Boolean, and as the ATiWith the ATl' consistent
When, Match (ATi,ATl')=1；When inconsistent, Match (ATi,ATl')=0；；
K is the quantity of the culture type label of the second culture activity, and g is the matching culture that the interest matching module obtains
The quantity of type label；Match(CTu,CTv') for u-th of culture type label C T in the second culture activityuWith it is described
L-th of matching culture type label C Tv' Boolean, and as the CTuWith the CTv' it is consistent when, Match (CTu,CTv')=
1；When inconsistent, Match (CTu,CTv')=0；
P is the quantity of the cultural keyword of the second culture activity, and q is that the matching culture that the interest coalignment obtains is closed
The quantity of keyword；KWxFor x-th of cultural keyword in the second culture activity, KWy' it is the cultural keyword of y-th of matching；
Sim is similarity function, and Sim ∈ (0,1], M is candidate's cultural activity recommended amount of the system default.
6. Public Culture information recommendation system according to claim 1, which is characterized in that the culture venue recommendation apparatus
Including venue matching module, venue temporal filtering module and venue travel time filtering module；
The venue matching module is configured to according to candidate's cultural activity that the cultural activity recommendation apparatus obtains and the text
Change venue information storehouse, match the cultural venue of candidate's cultural activity；
The venue temporal filtering module, be configured to according to the current time data, the cultural venue information storehouse with it is described
The cultural venue for candidate's cultural activity that venue matching module obtains is filtered, and obtains the first cultural venue；
The venue travel time filtering module is configured to according to the cultural venue information storehouse and the cultural geographical letter of venue
The distance time that computing device obtains is ceased, the first cultural venue that the venue temporal filtering module obtains is filtered, is obtained
Obtain goal culture venue and goal culture activity.
7. Public Culture information recommendation system according to claim 6, which is characterized in that the culture venue recommendation apparatus
Further include standing cultural activity recommending module；The standing cultural activity recommending module, is configured to obtain using collaborative filtering
User group similar with current user interest in the user behavior data storehouse is taken, and recommends user described in the target venue
The interested cultural activity of group is to active user.
8. Public Culture information recommendation system according to claim 7, which is characterized in that the standing cultural activity is recommended
Module includes user group acquiring unit, User Activity matrix construction unit, interest value computing unit and standing cultural activity and recommends
Unit；
The user group acquiring unit, be configured to according to the venue travel time filtering module obtain goal culture venue with
And the user behavior data storehouse, obtain the user group for once accessing the goal culture venue；
The User Activity matrix construction unit is configured to use with current according to the user group that the user group acquiring unit obtains
The cultural activity historical data at family builds user-active matrix；
The interest value computing unit is configured to, using LFM matrix decomposition methods, the User Activity matrix construction unit be obtained
User-active matrix calculate user's matrix and cultural activity matrix, and according to user's matrix and the cultural activity
Matrix computations go out interest value of the active user to each cultural activity；
The standing cultural activity recommendation unit is configured to the active user according to interest value computing unit acquisition to each
Cultural activity in the goal culture venue is carried out descending sort, and chooses the 1-N target by the interest value of cultural activity
Cultural activity in cultural venue is recommended, and wherein N is the standing cultural activity recommended amount of the system default, and N >=
1。
9. Public Culture information recommendation system according to claim 1, which is characterized in that the stroke recommendation apparatus includes
Traffic programme module and time planning module；
The traffic programme module is configured to the goal culture venue according to the cultural venue recommendation apparatus acquisition and utilizes institute
Cultural venue geography information computing device is stated, carries out the geo-location and path planning of venue, obtains traffic programme information；
The time planning module is configured to perform operations described below：
Step A1, the goal culture venue, the goal culture that obtain according to the cultural venue recommendation apparatus are movable with the traffic
The traffic programme information that planning module obtains, and when calculating venue distance by the cultural venue geography information computing device
Between；
Step A2, the time interval set according to the active user, the venue distance time and shown method according to the following formula
Obtain the total time that active user participates in goal culture activity：
t3=t1-t2
Wherein, t1For the time interval that the active user is set, t2For the venue distance time, t3It is participated in for active user
The total time of goal culture activity；
Step A3, according to the active user that the standing cultural activity recommendation unit obtains to the interest value of each cultural activity,
It calculates each goal culture activity and accounts for the time scale that the active user participates in the total time of goal culture activity；
Step A4 according to the time scale, the cultural venue information storehouse and the cultural activity information bank, is planned described every
The participation time of item target cultural activity, obtain time planning information.
10. according to Public Culture information recommendation system described in any one of claim 1-9, which is characterized in that described public
Cultural information commending system further includes feedback device；The feedback device is configured to receive the stroke as the active user
The information that the stroke recommendation apparatus is recommended is charged to the user behavior data storehouse by the information that recommendation apparatus is recommended.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711287156.2A CN108062366B (en) | 2017-12-07 | 2017-12-07 | Public culture information recommendation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711287156.2A CN108062366B (en) | 2017-12-07 | 2017-12-07 | Public culture information recommendation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108062366A true CN108062366A (en) | 2018-05-22 |
| CN108062366B CN108062366B (en) | 2020-08-11 |
Family
ID=62136280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711287156.2A Active CN108062366B (en) | 2017-12-07 | 2017-12-07 | Public culture information recommendation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108062366B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108898233A (en) * | 2018-06-15 | 2018-11-27 | 浙江浩逸文化发展股份有限公司 | Culture and arts exchange activity schedule method and system based on artificial intelligence |
| CN109886639A (en) * | 2018-12-28 | 2019-06-14 | 创泽智能机器人股份有限公司 | A kind of agenda managing method and system based on big data |
| CN110059177A (en) * | 2019-04-24 | 2019-07-26 | 南京传唱软件科技有限公司 | A kind of activity recommendation method and device based on user's portrait |
| CN112287243A (en) * | 2020-10-20 | 2021-01-29 | 山大地纬软件股份有限公司 | Service information recommendation device and method based on collaborative filtering algorithm |
| CN113742584A (en) * | 2021-08-31 | 2021-12-03 | 胡之乐 | Cultural tourism intelligent promotion method and system based on big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104834695A (en) * | 2015-04-24 | 2015-08-12 | 南京邮电大学 | Activity recommendation method based on user interest degree and geographic position |
| CN104933126A (en) * | 2015-06-11 | 2015-09-23 | 百度在线网络技术（北京）有限公司 | Activity recommending method based on electronic map, activity recommending device based on electronic map and activity recommending system based on electronic map |
| WO2017040224A1 (en) * | 2015-08-28 | 2017-03-09 | Microsoft Technology Licensing, Llc | Recommending a content curator |
| CN107122399A (en) * | 2017-03-16 | 2017-09-01 | 中国科学院自动化研究所 | Combined recommendation system based on Public Culture knowledge mapping platform |
-
2017
- 2017-12-07 CN CN201711287156.2A patent/CN108062366B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104834695A (en) * | 2015-04-24 | 2015-08-12 | 南京邮电大学 | Activity recommendation method based on user interest degree and geographic position |
| CN104933126A (en) * | 2015-06-11 | 2015-09-23 | 百度在线网络技术（北京）有限公司 | Activity recommending method based on electronic map, activity recommending device based on electronic map and activity recommending system based on electronic map |
| WO2017040224A1 (en) * | 2015-08-28 | 2017-03-09 | Microsoft Technology Licensing, Llc | Recommending a content curator |
| CN107122399A (en) * | 2017-03-16 | 2017-09-01 | 中国科学院自动化研究所 | Combined recommendation system based on Public Culture knowledge mapping platform |
Non-Patent Citations (1)
| Title |
|---|
| 董文: ""基于LDA和Word2Vec的推荐算法研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108898233A (en) * | 2018-06-15 | 2018-11-27 | 浙江浩逸文化发展股份有限公司 | Culture and arts exchange activity schedule method and system based on artificial intelligence |
| CN109886639A (en) * | 2018-12-28 | 2019-06-14 | 创泽智能机器人股份有限公司 | A kind of agenda managing method and system based on big data |
| CN110059177A (en) * | 2019-04-24 | 2019-07-26 | 南京传唱软件科技有限公司 | A kind of activity recommendation method and device based on user's portrait |
| CN110059177B (en) * | 2019-04-24 | 2023-09-26 | 南京传唱软件科技有限公司 | Activity recommendation method and device based on user portrait |
| CN112287243A (en) * | 2020-10-20 | 2021-01-29 | 山大地纬软件股份有限公司 | Service information recommendation device and method based on collaborative filtering algorithm |
| CN112287243B (en) * | 2020-10-20 | 2023-04-28 | 山大地纬软件股份有限公司 | Service information recommendation device and method based on collaborative filtering algorithm |
| CN113742584A (en) * | 2021-08-31 | 2021-12-03 | 胡之乐 | Cultural tourism intelligent promotion method and system based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108062366B (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Leonelli et al. | Data and society: A critical introduction | |
| CN108062366A (en) | Public culture information recommendation system | |
| Psorakis et al. | Inferring social structure from temporal data | |
| Yang et al. | Simulating land use change by integrating ANN-CA model and landscape pattern indices | |
| Moraru et al. | Using machine learning on sensor data | |
| CN109002492B (en) | Performance point prediction method based on LightGBM | |
| CN104834641B (en) | The processing method and related system of network media information | |
| CN106651392A (en) | Intelligent business location selection method, apparatus and system | |
| CN105760439B (en) | A kind of personage's cooccurrence relation map construction method based on specific behavior co-occurrence network | |
| Carvell et al. | Design and testing of a national pollinator and pollination monitoring framework | |
| Osojnik et al. | Tree-based methods for online multi-target regression | |
| CN107862022A (en) | Cultural resource commending system | |
| Li et al. | Knowledge transfer and adaptation for land-use simulation with a logistic cellular automaton | |
| Mehdipoor et al. | Developing a workflow to identify inconsistencies in volunteered geographic information: a phenological case study | |
| CN113139140A (en) | Tourist attraction recommendation method based on space-time perception GRU and combined with user relation preference | |
| De Groeve et al. | Individual Movement-Sequence Analysis Method (IM-SAM): characterizing spatio-temporal patterns of animal habitat use across landscapes | |
| Erhan et al. | Analyzing objective and subjective data in social sciences: Implications for smart cities | |
| Zhou et al. | Use of artificial neural networks for selective omission in updating road networks | |
| Giroux et al. | A high-frequency mobile phone data collection approach for research in social-environmental systems: applications in climate variability and food security in sub-Saharan Africa | |
| Wu et al. | Research themes of geographical information science during 1991–2020: a retrospective bibliometric analysis | |
| Zarzo‐Arias et al. | Importance of data selection and filtering in species distribution models: A case study on the Cantabrian brown bear | |
| Ramadiani et al. | Evaluation of student academic performance using e-learning with the association rules method and the importance of performance analysis | |
| Kachwala et al. | Predicting Rainfall from Historical Data Trends | |
| Bhole et al. | Smart site selection using machine learning | |
| Yu et al. | Modeling User Contextual Behavior Semantics with Geographical Influence for Point-Of-Interest Recommendation. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |