CN116997923A - Method for displaying places by using place similarity and travel duration - Google Patents
Method for displaying places by using place similarity and travel duration Download PDFInfo
- Publication number
- CN116997923A CN116997923A CN202180094592.6A CN202180094592A CN116997923A CN 116997923 A CN116997923 A CN 116997923A CN 202180094592 A CN202180094592 A CN 202180094592A CN 116997923 A CN116997923 A CN 116997923A
- Authority
- CN
- China
- Prior art keywords
- duration
- travel
- indication
- source
- sources
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title abstract description 157
- 238000004364 calculation method Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 description 15
- 238000003860 storage Methods 0.000 description 10
- 238000013459 approach Methods 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 9
- 238000005457 optimization Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000012546 transfer Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000012216 screening Methods 0.000 description 6
- 230000000670 limiting effect Effects 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 239000000446 fuel Substances 0.000 description 4
- 230000036961 partial effect Effects 0.000 description 4
- 230000007115 recruitment Effects 0.000 description 4
- 239000003795 chemical substances by application Substances 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012552 review Methods 0.000 description 3
- 229960001948 caffeine Drugs 0.000 description 2
- 230000001149 cognitive effect Effects 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011521 glass Substances 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000000399 orthopedic effect Effects 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- RYYVLZVUVIJVGH-UHFFFAOYSA-N trimethylxanthine Natural products CN1C(=O)N(C)C(=O)C2=C1N=CN2C RYYVLZVUVIJVGH-UHFFFAOYSA-N 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- VYZAMTAEIAYCRO-UHFFFAOYSA-N Chromium Chemical compound [Cr] VYZAMTAEIAYCRO-UHFFFAOYSA-N 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000009193 crawling Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003875 gradient-accelerated spectroscopy Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000007620 mathematical function Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 230000008433 psychological processes and functions Effects 0.000 description 1
- 230000002829 reductive effect Effects 0.000 description 1
- 238000013077 scoring method Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000009182 swimming Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000003442 weekly effect Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
An embodiment of the application is a method of determining a location indication using a location similarity and a travel duration. One embodiment is a method of operating on a real estate, which receives a commute destination. To improve utility, the method determines one real estate among a cluster of real estate having similar characteristics and similar commute times, and presents an indication of at least one real estate to a user.
Description
Technical Field
Cross Reference to Related Applications
The application is based on and claims priority date from the following applications:
and are incorporated by reference in their entirety.
Background
The present application relates to using similarities between locations to demonstrate the results of location searches or comparisons using travel durations within a traffic system.
When presenting the results of a search request to a user, the search engine typically sorts the results to reduce information overload for the user and to improve relevance. Two techniques are commonly used to sort information: clustering and scoring. The purpose of clustering is to group similar search results to avoid providing duplicate information to the user. The purpose of scoring is to rank the search results so that only the information most useful to the user is presented.
The concept of similarity has an intuitive meaning that there is sufficient similarity between data items. The term "similarity" is used in a broad sense, consistent with the interpretation of that term by one of ordinary skill in the art. Formally, the similarity can be modeled as a mathematical function that assigns a numerical value to a pair of data items in the range of 0 to 1. Wherein the number 0 indicates that the two data items are dissimilar and the number 1 indicates that they are similar, and the numbers between the two represent different degrees of approximate dissimilarity or approximate similarity. In one embodiment, two data items are defined as similar when the number is at least one threshold, for example at least 0.9. Any data item is considered similar to itself in the present disclosure. In a specific context, the similarity is defined in various ways according to the specific case. In one embodiment, the definition is made using text: for example, when at least 90% of the text portions of two web pages match, such as an n-grams algorithm, where n=5, then a similar web page is defined. In one embodiment, the definition is made using numerical values: for example, two real estate sources are defined as similar when their prices differ by less than 5% and they are located in the same geographic location. In one embodiment, the similarity is defined using artificial intelligence software executed on the data items by a computer system. Examples of artificial intelligence software include: neural networks, support vector machines, markov models, bayesian networks, and the like. For example, an artificial intelligence software is used to define the similarity between two real estate sources, wherein the software is executed on images associated with the real estate sources, resulting in a similarity value in the range of 0 to 1. In one embodiment, the similarity is defined on normalized data items, such as: the text "San Francisco, cali" contained in the data item is converted to the text "San Francisco, CA", the area in square feet is converted to the area in square meters, or the pixel color of the image is readjusted to achieve an average luminance of 50%. In one embodiment, the similarity is made using the distance between the data item represented by the mathematical vector and the vector, e.g. chebyshev distance, minkowski distance, etc. In one embodiment, the similarity is defined as: cosine similarity, string similarity (e.g., lycenstant distance (Levenshtein distance)), semantic similarity, etc. In one embodiment, the coordinates of the vector are normalized, e.g., with a mean of 0 and a variance of 1. In one embodiment, the similarity is defined using a combination of at least two similarities, for example: text contained in the data item is matched but the image contained in the data item is used with artificial intelligence software and the two results are combined, for example using a weighted sum. In one embodiment, the similarity focuses only on a portion of the data item, such as ignoring mortgage information for the real estate source. It will be apparent to those of ordinary skill in the art that many other ways of defining similarities are possible without departing from the scope and spirit of the present invention.
The clustering problem has been widely studied. For example, please refer to the prior art Review survey of Jain, murty and Flynn ("Data Clustering Review (Data Clustering: A Review)", "ACM Computing Surveys", volume 31, stage 3, month 9 1999). In short, given a certain number of data items and a method for describing the similarity between data items, the goal is to assign these data items to groups of similar data items. The prior art has developed a number of clustering methods, such as connection-based clustering, e.g. hierarchical clustering; center-based clusters, such as K-means (K-means) clusters; distribution-based clustering such as expectation-maximization algorithm; density-based clustering, such as DBSCAN; grid-based clustering such as STING or CLIQUE; pre-clustering, such as Canopy clustering; subspace clustering, such as CLIQUE or sullu; projection clustering, such as PreDeCon; etc. In one embodiment, the clustering method calculates clusters that meet additional requirements. Examples include requirements of minimum or maximum cluster size, minimum or maximum value of cumulative similarity within a cluster, etc. The additional requirements are determined based on the context of the usage clusters.
One basic clustering method involves calculating the similarity between each pair of data items and assigning them to the same group when the similarity reaches at least a certain threshold. However, when processing large amounts of data, the quadratic increase in the logarithm of the data item results in this basic approach not being practical. To overcome the scalability problem caused by the quadratic growth of the logarithm of data items, search engines typically employ a heuristic approach to filter pairs of data items that are unlikely to have similarity. For example, prior art US 6658423 B1 describes such a heuristic method. In this case, each data item is a web page. The heuristic assigns a hash value to each web page, where the hash value may be considered as a very short text derived from potentially very long web page text. The web pages are then grouped according to hash values (which may be accomplished simply by sorting, binning, etc. the hash values) and similarity is calculated only between web pages having the same hash value. The hash value is generated in such a way that the hash values of the two web pages match, i.e. are equivalent to the two web pages having a similarity. This may be achieved, for example, by using an n-gram. Thus, the heuristic method can significantly reduce the number of similarities that need to be calculated compared to the underlying quadratic method.
In order to enable practical clustering in certain application areas, several other heuristics have been developed in the prior art. For example, where the data item is real estate, the heuristic methods in the prior art include the documents US 20150012335A 1, US 9858528 B2 and US 10776888B 1. And when the data item is a recruitment advertisement, heuristic methods in the prior art comprise the following documents, namely US 10043157B 2 and Burk, javed and Balaji's paper "Apollo: near-Duplicate Detection for Job Ads in the Online Recruitment Domain", 2017 International data mining seminar (International Conference on Data Mining Workshops 2017).
Many scoring methods have been developed in the prior art. For example, please refer to prior art US 7058628B1, which scores specific areas of web search engines based on PageRank; and please refer to the prior art US 7974930 B2, which scores a particular field of real estate based on real estate characteristics.
With reference to prior art WO 2019164727, recent advances in navigation technology have created an engine that searches or compares real estate using commute time. For example, if a user requests to specify a workplace, these techniques can quickly determine the exact duration of travel between each property and the workplace within the metropolitan area. Thus, a deep search of the real estate market can be performed. However, prior art approaches may not enable presentation of search results in a more practical manner. This presentation needs to solve the problem of avoiding duplicate information and improving the relevance of the search results, but must be done in a practical and scalable way. The invention discloses a method for achieving the aim.
Disclosure of Invention
An overview of the present invention is presented herein to simplify one way to allow the reader to understand certain aspects of the claimed subject matter. This summary is not an extensive overview of the invention, and is intended to neither identify key or critical elements of the invention nor delineate the scope of the invention. The purpose of this overview is to summarize some concepts in a form that is easier for one skilled in the art to read. The reader should see the disclosure of the invention for details. The reader should refer to this disclosure for details.
Specific embodiments of the invention are as follows:
1. a method of determining an indication of a plurality of locations within a traffic system using a trip length and a similarity, the method characterized by:
(a) Receiving a request included in at least one place within the traffic system;
(b) Determining at least two isochrone locations contained in the plurality of locations, wherein a travel length between each isochrone location and the at least one location within the transportation system is contained within a range;
(c) The determination is made using one of the following steps
i. Determining a plurality of similar places included in the at least two isochronal places, and determining the indication of the plurality of similar places; or alternatively
Selecting at least one first location that is dissimilar from at least one second location and that is contained in both the at least two isochronal locations, and determining an indication of the at least one first location and the at least one second location;
and
(d) Responding to the request with the indication.
2. A method of determining an overview of a plurality of locations within a traffic system using a trip length and a quantity, the method characterized by:
(a) Receiving a request included in at least one place within the traffic system;
(b) The calculation is included in theIn a sequence of two or more places, wherein,
i. for the first location and the second location comprised in the sequence,
travel length within the traffic system between the first location and the at least one location
And (3) with
At least one distance between the second location and the at least one location between "travel length within the traffic system", and
so long as the travel length between the fourth location and the at least one location is within the traffic system
Including the vicinity of the "length of travel within the transit system" between the third location and the at least one location,
Then
The number associated with the third location included in the sequence is at most equal to the number associated with the fourth location included in the plurality of locations;
(c) Determining the overview comprising the sequence indication; and
(d) Responding to the request with the overview.
3. A method of determining an indication of at least two alternatives among a plurality of points of interest within a transportation system, the method characterized by:
(a) Receiving a request including a location within the transportation system;
(b) Determining the at least two alternatives, wherein,
the length of travel between each alternative and the location within the traffic system is within a threshold of a shortest travel;
(c) Determining the indication of the at least two alternatives, wherein the indication is a non-unitary and non-stroked description; and
(d) Responding to the request with the indication.
4. A method of determining an indication of at least two locations within a traffic system using a length of travel and a length of travel estimated, the method characterized by:
(a) Receiving a request included in at least one place within the traffic system;
(b) At least two estimated stroke lengths are determined, wherein,
The at least two estimated travel lengths include an estimated travel length within the traffic system between the at least one location and each of the at least two locations;
(c) Selecting one or more of the at least two locations using the at least two estimated travel lengths, wherein the number of the one or more locations is at most a predetermined limit;
(d) Determining at least one travel length, the at least one travel length comprising a travel length within the traffic system between the "each of the one or more locations" and the "at least one location";
(e) Determining the indication of the one or more locations using the at least one run length; and
(f) Responding to the request with the indication.
Embodiments of the invention also include computer systems and devices implementing any of the methods described above.
The embodiments described in this disclosure are for illustrative purposes only and are not intended to limit the present invention. It will be apparent to those skilled in the art that various modifications and variations can be made to the embodiments without departing from the scope or spirit of the invention.
In this specification, the terms "first," "second," "said," and the like are not used in any limiting sense, but are used for distinguishing purposes unless otherwise expressly expressed from the context. The expression in the singular encompasses the plural unless the context clearly indicates otherwise. The terms "having," "including," "containing," and similar terms, mean the presence of a component or feature, and do not exclude the presence or addition of other components or features.
Drawings
The accompanying drawings of the present disclosure illustrate various features and advantages of certain embodiments of the invention:
FIG. 1 depicts an example of an engine internal data flow for searching or comparing real estate using commute time;
FIG. 2 depicts an example cluster of real estate sources contained within an isochrone;
FIG. 3 depicts an example user interface for receiving a user request and displaying a response to the user;
FIG. 4 depicts an example of an overview contained in a user interface;
FIG. 5 depicts example steps contained in a user interface for determining alternative school directives;
FIG. 6 depicts example steps of a two-stage method of determining a trip duration.
The drawings are for illustrative purposes only. It will be apparent to those of ordinary skill in the art that many more drawings can be made from these drawings without departing from the scope and spirit of the invention.
DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION
The present invention relates to the general case of showing places using similarities between places and travel duration. However, for ease of explanation, we first illustrate the invention by using one embodiment of the engine that searches or compares real estate with commute time. For brevity we simply refer to the engine. This example is not limiting. In the following sections, we will explain how this method works in a general case.
1 exemplary embodiment
We describe an exemplary embodiment of the present invention. In the description, we use the term "module". As is well known in the art, the term "module" refers to a computer system (and thus may be considered a computer subsystem) that provides some specific functionality. In one embodiment, the engine is divided into three modules, (1) an acquisition module, (2) an indexing module, and (3) a service module. The division of specific modules for an engine is only exemplary and not mandatory. It will be apparent to those of ordinary skill in the art that the engine may be divided into modules in other ways without departing from the scope and spirit of the embodiments.
We will next describe the function of the module. In the description, we refer to the schematic diagrams of fig. 1, 2, 3, 4, 5 and 6, wherein prime notation denotes at least one example of the recited element. We describe the functions as a series of steps that occur in a particular order. However, this order is not limiting. It will be apparent to those of ordinary skill in the art that certain steps may be performed in other sequences, partially concurrently, and combined or omitted without departing from the scope and spirit of the embodiments.
Acquisition of
The acquisition module (1002) acquires information about real estate sources from at least one source (1001). These sources can be categorized into a number of types including, but not limited to, the following: (1) A direct source corresponding to a property intermediary (broker), landlord, construction company, etc. inputting property source characteristics to the collection module; or (2) indirect sources, websites, smart phone applications, etc. corresponding to the release of the property source features, which are crawled by the acquisition module. The information is obtained from the first information source in the following ways: conventional call centers, websites, smart phone applications, etc., or similar user interfaces having features that allow their user to enter. The information is obtained from the second information source by: the computer system accesses information sources, such as web sites or smart phone applications, via the internet, which themselves publish features. During the visit, features will be crawled. This process is commonly referred to as data crawling. Characteristics of the real estate source include, but are not limited to, at least one of: a name; an address; a geographic location; a time node of a property to be marketed or returned to the market; a floor; the number of floors of the building; the geographic orientation of the door; the geographic direction of the window; characterization of the scene outside the window; price and amount; a lease amount; deposit amount; borrower information; loan information; mortgage loan information; month management fees; the month management fee covers the description of the content; a date of check-in; square meter area; land square meter area; price per unit area, such as per square meter; the number of bedrooms; number of bathrooms; structural or layout features; heating or cooling modes; an elevator description; parking lot instructions; swimming pool description; garden or backyard instructions; a child playground instruction; a gymnasium description; literal description of property by the homeowner or property broker; pictures of indoor, outdoor or outdoor scenery; indoor tour film; sound or noise recorded through an opened window; an air quality measurement; descriptions of intermediaries or brokers; ratings of intermediaries or brokers; or an identifier assigned to the real estate source by the information source. In one embodiment, the acquisition module may also acquire point of interest information, which are related to real estate sources by travel, and both are contained in a transportation system. Any point of interest is an arbitrary location. The points of interest and example information thereof include, but are not limited to, at least one of: schools (example information: school type (public, private, career, elementary school, junior middle school, high school, university, etc.), ranks of schools in other schools, school fees, recording criteria, recording probability, or geographical range of the area where schools are located), workplaces (example information: job description or salary information) or convenience facilities (e.g., public transportation sites, highway entrances, parking lots, senior activity centers, parks, hospitals, clinics, pharmacies, restaurants, shops, convenience stores, laundry, banks, automated teller machines, government offices, crime reports, police stations, or military bases). In one embodiment, the information acquired by the acquisition module is stored in a non-volatile storage medium (e.g., a database). In one embodiment, the process of collecting data may be interpreted as a measurement of the real world, where the measurement obtains information about the physically present entity. The physical presence may be in the form of data about the entity stored by the source in a non-volatile storage medium. It will be apparent to those of ordinary skill in the art that the source of the real estate could be obtained in many other ways without departing from the scope and spirit of the embodiments of the present invention.
In one embodiment, the acquisition module is continuously operated, thereby creating a real estate market view that evolves over time. At some point in time, there may be multiple room sources associated with the actual house, for example, because the room sources are from multiple sources. Such multiple information may contradict or conflict, for example, two different real estate agents have entered two different square meter areas. The room source information created at a certain moment may be modified later, e.g. the homeowner edited a room property description, uploaded a new room property picture, etc. The source information may be deleted or restored, for example, when a potential tenant cancels a lease. The source information may be intentionally ambiguous, for example, when the landlord wishes to hide private information, an approximate range of the source geographic location may be provided, for example within 100 meters of a square circle, or a range of floors within a building, for example "high floors". It is also possible that a source of a house may be out of date, for example, when a real estate has been sold, but the sales information has not been reflected on any source. False sources of property that fool potential tenants, or malicious sources of property that deliberately skew features of the sources of property, may also occur.
1.2 index
The indexing module (1003) normalizes the data obtained by the acquisition module to a form storable in a non-volatile storage medium. In one embodiment, this form represents the real estate source as a feature vector, with each feature being associated with a value. The data type of a value includes, but is not limited to, any of the following: text, numbers, images, video, sound or smell. A value is encoded in a suitable manner into a computer-accessible form. For example, a price range may be represented as two features and numerically: high and low cost. In one embodiment, the non-numeric value is mapped to a number. For example, the feature "elevator" has two possible values: "available" and "unavailable", the two values being mapped to 1 and 0, respectively. In one embodiment, the generated features and their values are added to the feature vector. In one embodiment, we add a new feature and a value for each non-numeric feature. In one embodiment, one feature and its value are calculated using at least one other feature and its value. For example, the price value is divided by the area value to calculate the unit price, or the value is calculated by artificial intelligence software executed on the features and values by a computer system. In one embodiment, the value of a characteristic is a time node representing the time at which the acquisition module acquired a particular real estate source. In one embodiment, the feature vectors are sparse, meaning that some real estate sources may lack certain features, but other real estate sources may have certain features. For example, not every real estate source has pictures of a bathroom. In one embodiment, the feature vector is represented as a hash map or list. In one embodiment, the indexing module reconciles conflicting information, for example, using a majority voting method, an averaging method, or the like. It will be apparent to those of ordinary skill in the art that many other methods of data normalization may be employed without departing from the scope and spirit of the present embodiments.
Normalization is implemented as a module that is tailored to a particular data source. The modules are created manually by studying the data sources in order to properly interpret the meaning of the data obtained from the data sources. Once completed, the module will run automatically, without manual supervision.
In one embodiment, the indexing module creates an inverted index of at least one real estate source (1005). Subsequently, during request processing, the pre-computed inverted index can quickly identify the property source for which a particular feature has a particular value. For example, by reverse indexing, all real estate sources with exactly 3 bedrooms can be quickly identified. In one embodiment, the manner in which the inverted index is implemented includes: hash graphs, ordered lists, etc. In one embodiment, the indexing module may build an inverted index for any feature, e.g., may build an inverted index for each feature that is frequently present in a user request, e.g., such as: price, number of bedrooms, number of bathrooms, square meter area, etc. It will be apparent to those of ordinary skill in the art that many other methods of creating at least one inverted index can be used without departing from the scope and spirit of the present invention.
In one embodiment, the indexing module establishes at least one real estate source cluster (1006). Pre-computing clusters is useful because they help identify similar house sources during request processing. Clustering uses a similarity concept, such as one of the similarity concepts mentioned in the present disclosure. In one embodiment, the module clusters real estate sources using geographic distances. For example, the module reads feature vectors associated with real estate sources from a non-volatile storage medium. The geographic locations represented in the feature vectors are clustered using a greedy approach. For example, sites are processed in any order, if a location cannot be added to any previously created clusters, a new single cluster is created without exceeding the cluster radius (e.g., set to 10 meters). The real estate sources are then assigned to clusters according to the geographic location clusters. In one embodiment, the indexing module builds clusters using defined similarities over any feature, e.g., over each feature that is frequently present in user requests. In one embodiment, the indexing module uses the similarity of two or more features to establish clusters, e.g., consider only house sources that match in both price and area to be considered similar. It will be apparent to those of ordinary skill in the art that many other methods of establishing at least one cluster may be employed without departing from the scope and spirit of the present embodiments.
In one embodiment, the inverted index or clustering results are limited to room sources obtained by the acquisition module over a period of time, such as the last 24 hours. In one embodiment, the period of time may be relatively broad, such as years, in order to identify numerical trends over time. In one embodiment, the inverted index or clustering result is limited to only the house sources corresponding to the property on the market.
In one embodiment, the indexing module may also operate on points of interest in a manner similar to that described above for real estate sources.
In one embodiment, the normalized data, inverted index, or clustering result is saved to a non-volatile storage medium.
In one embodiment, the indexing module is run continuously, maintaining a view of normalized data, inverted indexes, and clustering results over time.
1.3 service
The service module is configured to receive a request, generate an indication, and respond to the request with the indication. Next we will describe several embodiments.
1.3.1 indication
The service module (1008) is configured to receive a request (1007) from a user. Wherein one request contains an explanation of at least one commute destination (2001) (3002). The commute destination is an arbitrary location, such as: geographic location, address, or point of interest. The request also includes parameters for determining a manner in which the at least one commute destination forms the at least one commute path. A commute path refers to a sequence of trips between pairs of endpoint positions. The commute path starts or ends with a real estate source and includes a commute destination. The commute path may take a variety of forms, such as: the round-trip commute path (e.g., go home, then work, then home, or go home, then leave the closest school, then home), the open commute path (e.g., go home, then school in school, then piano class, last home), or the discontinuous commute path (e.g., go home, then work, then other homes). It will be apparent to those of ordinary skill in the art that other ways of forming the commute path may be used without departing from the scope and spirit of the invention. Other parameters may also be included in the request, such as: departure time, arrival deadline, probability of arrival before deadline, travel mode (car, bus, subway, walking, combined trip, etc.), or travel frequency. For example, a workplace visit 5 times per week, a school visit 3 times per week. Other parameters may also include screening limitations, i.e. limitations on travel, such as: what kind of vehicle should be used when the journey is performed; upper limit of transfer times; allowed transfer types (e.g., subway-bus transfer, or bus-subway transfer); a time window for transfer; limitation of walking duration; or a limitation on points of interest that may be included in the commute path (e.g., only the top 10% of schools). Given a particular real estate, we can determine the duration of travel of at least one commute path. We will simply refer to the result as: the duration of travel between the real estate and the at least one commute destination. In one embodiment, the duration of the trip between the real estate and the at least one commute destination is a value that reflects the total amount of time that all residents of the real estate spend on the trip over a period of time (e.g., daily, weekly, monthly, etc.). These residents may be: families, roommates, office associates, etc. living in the real estate. Other parameters may include financial information of the user, such as: financial property reports, credit ratings, revenue information (e.g., such as time payroll, job category, job payroll information, mortgage application); or school information of the user, such as: school entrance questionnaires, such as those containing math exams. The request may also contain a description (3001) of the desired characteristics of the property or its value, for example, an area in the range of 80 to 90 square meters, or the desired characteristics of the point of interest or its value. It will be apparent to those of ordinary skill in the art that many other forms of requests are possible without departing from the scope and spirit of the invention.
The service module then identifies a real estate source L matching the request parameters (2002). For example, when the desired feature is specified as "between 80 and 90 square meters in area," the module identifies all sources of room having an area between 80 and 90 square meters. In one embodiment, the identification process uses an inverted index built by an indexing module, for example, by intersecting sets of room source identifiers, each associated with a desired feature and its value specified in the request. In one embodiment, the identification is limited to only the room sources acquired by the acquisition module over a period of time, such as the last 24 hours. In one embodiment, identification is limited to only the house sources corresponding to the property on the market. In one embodiment, L includes all real estate sources. It will be apparent to those of ordinary skill in the art that many other methods of identifying matching real estate sources may be employed without departing from the scope and spirit of the present invention.
The service module determines at least one trip duration between each geographic location of the house source L and at least one commute destination using any means known to those of ordinary skill in the art, for example using the prior art navigation service mentioned in this disclosure (1010). For example, for a particular geographic location H of the real estate source, a particular geographic location of a workplace, and a particular geographic location of a school, the navigation service calculates a travel duration D from H to workplace and back to H w And a trip duration D from H to school back to H s . Then, the travel duration between the real estate source and the at least one commute destination is D w +D s Including D w And D s Is contributed by Buddha year. In one embodiment, the travel duration is derived from other travel durations. For example, the trip duration is a weighted sum calculated based on the trip frequency, e.g., (5D) w +3·D s )/(5+3). In another example, the travel duration is the difference between travel durations, e.g., the travel duration of the current residence (included in the request) minus the travel duration of the candidate new residence (included in the house source L). In one embodiment, the trip duration is the shortest trip duration, e.g., calculated using Dijkstra's algorithm on a graph of a simulated traffic system, the graph constructed using the prior art method disclosed herein. In one embodiment, the trip duration is an estimated trip duration, e.g., differing from the shortest trip duration by a multiplicative factor or an additive sum, e.g., a factor of 2 or one and 15 minutes or 1000 meters (the description of such estimated trip duration corresponds to a trip duration calculated using the methods described in the present disclosure or the prior art referred to in the present disclosure). In one embodiment, the navigation service determines the travel duration using any of the methods described in the following prior art:
(a) The representative position, i.e. the position that frequently occurs in the shortest travel, as described in WO2019164727, may be defined as: positions included in a traffic system, wherein the number of representative positions is at most the size of the traffic system multiplied by a predetermined ratio of at most 1, examples of representative positions include:
(i) Landmarks, portals, hinges, beacons, seeds, traffic nodes, etc., please refer to Sommer's "short-Path Queries in Static Networks", ACM Computing Surveys, volume 46 (4) 2014;
(ii) Transfer stations, and global sites where transfer may occur during long distance connections, as described in US 8756014B 2;
(iii) The center of the grid as described in CN 105975627A; or (b)
(iv) Boundary vertices as described in US 9222791 B2;
(b) Processed map data, including nodes representing pre-screened map features, as described in US 9250075 B2;
(c) A simplified road map as described in US 91959593 B2;
(d) A hierarchy of polygon layers as described in US 7953548 B2;
(e) Grid, as described in CN 105975627A;
(f) A sub-graph obtained by excluding at least one waypoint, as described in US 8949028 B1;
(g) One or more fixed nodes as described in EP 2757504 A1;
(h) A forward part-path and a backward part-path as described in EP 1939590 B1;
(i) An overlay as described in US 9222791 B2;
(j) Intermediate waypoints as described in US20110251789 A1; or (b)
(k) As described in US 8477209 B2, US 8738286 B2, US 8756014 B2, US10533865 B2, KR 101692501B1 or CN 104240163a, transmissions are made between a source station as or adjacent to a source location and a destination station as or adjacent to a destination location;
or any known technique, including:
(l) Shrink hierarchies, such as Geisberger, sanders, schultes and belong works: "Contraction Hierarchies: faster and Simpler Hierarchical Routing in Road Networks", "society of laboratory and efficient algorithms in 2008 (Workshop on Experimental and Efficient Algorithms 2008)", or Belling, goldberg and Werneck: faster Batched Shortest Paths in Road Networks, a society of traffic modeling, optimization and systematic algorithm methods in 2011 (Workshop on Algorithmic Approaches for Transportation Modeling, optimization, and Systems 2011);
(m) GRP, GRASP and PHAST based technologies, such as Baum, buchhold, dibbelt and Wagner works: fast Exact Computation of Isocontours in Road Networks, ACM Journal of Experimental Algorithmics, month 10 of 2019;
(n) techniques listed in Sommer's survey report: "Shortest-Path Queries in Static Networks", "ACM Computing Surveys", volume 46 (4), 2014; or (b)
(o) Bast, belling, goldberg, miiller-Hannemann, pajor, sanders, wagner and Werneck, techniques listed in the survey report: "Route Planning in Transportation Networks", "Algorithm Engineering" 2016. In one embodiment, the service module determines a trip duration for the geographic location of the house source prior to identifying L, stores the trip duration in a non-volatile computer-readable storage medium, and retrieves the trip duration from the storage medium after identifying the location in L. In one embodiment, the service module determines a travel duration for a subset of the room sources, e.g., only the travel duration of the room sources in the vicinity of the metropolitan area, e.g., the travel duration of the room sources within a travel radius of 500 meters or 1 minute from the preset location. Various methods of selecting a subset are described in the present disclosure. It will be apparent to those of ordinary skill in the art that many other methods of determining the duration of travel may be used without departing from the scope and spirit of embodiments of the present invention.
The service module then groups house sources L using the neighborhood of trip durations. The groupings relate to the concept of isochrones, which are lines on a map that connect points with the same travel duration. However, we use the concept of isochrones (2003) in a broad sense, which represents a range in a broad sense. In one embodiment, the range is set to a shorter travel duration, for example 15 minutes. In one embodiment, the range is set to a small distance, such as 500 meters, for example. In one embodiment, the range includes at least two values. For example, the service module considers a continuous time range of width M minutes: i.e., [0, M), [ M, 2M), [2M, 3M), etc., and so on, e.g., M is set to 15. One group i is associated with the range [ iM, (i+1) M). Then, when D is the duration of the trip between the geographical location of the house source and at least one commute destination, then the house source is assigned to a group i whose association range includes D. In another example, a group consists of a limited number of sources that appear in succession in the source ordered by travel duration, e.g., up to 1000 sources per group. In this case, the ranges may have different widths. In one embodiment, a group includes real estate sources without any geographic restrictions, such as, for example, a house driven eastward 1 hour from a work site and a house driven westward 1 hour from a work site (2004). In one embodiment, a group includes real estate sources within a certain community (2005) limited to a large area of the venue, such as sources within a 500 meter or 1 minute range from a preset location. In one embodiment, certain ranges are overlapping. In one embodiment, a packet is calculated corresponding to an timeline that includes a range of travel durations, such as a predetermined range or a range specified in the request. It will be apparent to those of ordinary skill in the art that many other methods of grouping house sources using contiguous areas of travel duration may be employed without departing from the scope and spirit of embodiments of the present invention.
The service module then determines clusters of similar house sources in any one group (2004) (2005). In one embodiment, the service module uses any of the similarity concepts mentioned in this disclosure. In one embodiment, the clusters are determined using any of the clustering methods disclosed herein. One cluster may contain only one room source, for example when there are no other similar room sources, or one cluster may contain multiple room sources. In one embodiment, the similarity may be affected by the user, for example, the user's request may be: "ignore multiple intermediaries and agents", in which case the house sources displaying different agents but with similar advertising content would be considered similar house sources. In one embodiment, pre-computed clusters are used to accelerate clustering during request processing, e.g., pre-clustering house sources using text features using pre-computed clusters, and then clustering each pre-cluster using user-affected similarity. It will be apparent to those of ordinary skill in the art that many other methods of determining clusters can be used without departing from the scope and spirit of the invention.
The service module determines a score for a room source or a cluster of room sources. The score is an entity that can be compared to other scores to determine the order. For example, the score is a numerical value that is compared using "greater than. In one embodiment, the score of the house source uses at least one feature and its value, examples include: travel characteristics (e.g., geographic location of a house source and duration of travel between at least one commute destination, etc.), point of interest characteristics (e.g., characteristics and values of points of interest contained in at least one commute destination, such as a rank of a school), location characteristics (e.g., crime occurrence, availability of a school, local services, convenience facilities, etc.), financial characteristics (e.g., market standard value, price up or down, etc.), reputation characteristics (e.g., past rental performance of a real estate intermediary, feedback of a previous buyer, etc.), and temporal characteristics (e.g., number of days on the market of a listing house source, etc.). In one embodiment, the score is defined as the "characteristics" listed in prior art US7974930B 2. In one embodiment, the scores are calculated by artificial intelligence software executed by a computer system that provides a realistic basis for predictive scoring based on past user interaction training with the service module, such as clicking on network links related to features, values, and requests. In one embodiment, the score is determined using any mathematical formula. For example, the score of a house source is a weighted sum of its numerical eigenvalues. The weights may be positive, negative or zero. In one embodiment, we use equal weights. In one embodiment, the weights are determined so that each feature has an equal chance impact score, such as: for features with higher median values in all house sources, we use lower weights. In one embodiment, we use weights that can highlight specific features, such as: we use higher weights for financial features or higher weights for newer house sources. In one embodiment, the user may have an impact on the score, for example, the user may require: "order by build date" or "prioritize schools top" in which case the score for the relevant house source will be increased, for example, by appropriate adjustment of the weights. In one embodiment, the scores of the house source clusters are mathematical statistics of house source scores in the clusters, such as: the highest score of any room source in the cluster, or a weighted sum of the digital signature values in the cluster. In one embodiment, the score is a vector whose coordinates are calculated using at least one feature and its value, for example: a two-dimensional vector whose first coordinates are negative of the duration of the stroke and whose second coordinates are calculated using the above-mentioned features and their values. In one embodiment, such vectors are ordered in a lexicographic order. In one embodiment, the score is alphabetically ordered text, such as the name of an apartment community. It will be apparent to those of ordinary skill in the art that many other methods of determining the score may be used without departing from the scope and spirit of the present embodiments.
The service module selects clusters. In one embodiment, the module selects clusters that score highest, e.g., the first 20 clusters. In one embodiment, the module selects a number of clusters that meets at least one additional requirement. In one embodiment, an additional requirement is to select clusters within a range of travel durations, such as travel duration range [0, M ]. In one embodiment, an additional requirement is that a certain number of clusters is selected at most, for example 5 clusters at most. In one embodiment, the additional requirement is the geographic sparsity of the selected clusters, e.g., a preset number of clusters in any neighborhood is selected at most. For example: the clusters are processed in the order of the score using a greedy method, with the highest score being preferred, and if a cluster is selected that has a neighborhood of a certain cluster that exceeds a threshold, for example, more than 2 clusters are selected within a 500 meter or 3 minute path, the clusters are prevented from being selected. In one embodiment, the additional requirement is that the selected clusters have different eigenvalues. For example, a set of two-living apartments and a set of three-living apartments are selected. In one embodiment, clusters are processed by a greedy method and clusters whose specific eigenvalues are similar to those in the selected clusters are selected and excluded, thereby ensuring diversity. In one embodiment, the additional requirements are specified in the request. For example, the request may require a room source within a particular school district. In one embodiment, the module selects clusters using any of the clustering methods disclosed herein, wherein the clustering method operates on rows of data items, each of which is a cluster. For example, the clustering method determines a data item as the centroid, which becomes a selected cluster. It will be apparent to those of ordinary skill in the art that many other methods of selecting clusters can be used without departing from the scope and spirit of the present embodiments.
In one embodiment, the service module determines an indication of at least one room source in the cluster and responds to the request with the indication (1009). The clusters are determined using similarities between the house sources and travel durations between the house source geographic location and at least one commute destination, e.g., as described above. The indication includes, but is not limited to, at least one of:
(a) The house sources with the highest scores in the clusters;
(b) The location of the house source in the cluster relative to the score (3007);
(c) Travel duration of the house source (3003), (3004);
(d) A portion of the journey duration, such as the walking duration in a public transportation journey;
(e) An excerpt of the house source (3005), i.e. a literal representation of the house source;
(f) A web link (3006) of a house source published by a source;
(g) At least one characteristic of the source and its value (3005);
(h) Mathematical statistics of values, such as (i) histograms of eigenvalues, (ii) frequency statistics, such as most frequent eigenvalues or least frequent eigenvalues, (iii) random samples of eigenvalues, (iv) maximum or minimum values of values, (v) average, median, percentile, standard deviation or variance of values, or (vi) fractions of values within a certain range, such as below or above a threshold;
(i) Based on mathematical statistics of a period of time, such as price trends over the last 5 years;
(j) Excerpt of any house source in the cluster;
(k) Excerpts of at least two house sources in a cluster, such as: (i) An indication of the number of sources (3007), or (ii) a difference between two sources, such as showing that one of the sources is cheaper, or showing that one of the sources is published by a more trusted intermediary;
(l) A composite room source constructed by combining features of at least two room sources in a cluster and their values, for example: displaying telephone numbers of all property intermediaries advertising the specific house, but displaying the bedroom number of the specific house only once;
(m) a graphical or textual representation of the clusters; or (b)
(n) corresponds to any of the above, but involves one point of interest in at least one commute destination.
In one embodiment, an indication of the selected clusters is determined, wherein the indication includes an indication of at least one room source for each selected cluster. In one embodiment, the indication of any one set of real estate sources is determined in the manner described above for determining the indication of at least one real estate source in the cluster. For example, including mathematical statistics of the price of the house source currently on the market, or of the price of the house source that matches the user's request. It will be apparent to those of ordinary skill in the art that many other methods of determining the indication may be used without departing from the scope and spirit of the present embodiments.
In one embodiment, we explicitly restrict the indication of at least one room source (e.g., the indication of all room sources contained in the isochrone) such that the indication does not include any information other than the explicitly listed information. These limitations include any indication of at least one of the above-mentioned sources, but with the addition of the qualifier "only", e.g
(a) Only contains the house source excerpt with the highest score in at least one house source;
(b) There are only two combinations of indicators; or (b)
(c) When k is equal to or greater than 3, only k combinations of indexes are displayed.
This has an advantage in that these restrictions reduce the information burden on the user and increase the relevance. For example, only one is displayed in the house source for all journey durations between 10 minutes and 20 minutes, which gives the user little cognitive load while providing useful information.
In one embodiment, the service module responds to the request (1009), the indication indicating that at least two clusters are dissimilar. For example, the method may select two clusters that are at least a threshold distance apart. The concept of inter-cluster distance is known in the art, for example: the minimum distance between any pair of house sources in the two clusters, or the distance between the centroids of the two clusters. For example, the threshold is a stroke of 1000 meters or 1 minute. The method then determines an indication of the two clusters, which facilitates diversified responses to the request. For example, as described above, the similarity between the house sources and the travel duration between the house source geographic location and the at least one commute destination are utilized to determine the at least two clusters. Thus, a first room source in one cluster may be considered dissimilar to a second room source in another cluster. In one embodiment, the travel duration is within a range, but in another embodiment the travel duration is not required to be within a range. It will be apparent to those of ordinary skill in the art that many other methods of responding to the indication of at least two clusters may be used without departing from the scope and spirit of the present embodiments.
1.3.2 overview
In one embodiment, the method uses the number and trip duration associated with each source to determine the real estate source overview. One advantage of the overview is that it can help the user find a source of room that balances individualization between quantity and duration of the journey. We describe an embodiment in which the quantity is interpreted as a sales price. This explanation is not limiting. It will be apparent to those skilled in the art that other explanations may be made to the number without departing from the scope and spirit of the present embodiment. To simplify the description of how the overview is determined, we will refer to a single house source. However, it should be understood that each such individual source may correspond to a similar source cluster determined in accordance with the methods disclosed herein. Each travel duration mentioned in the description refers to a travel duration between the real estate and at least one commute destination.
In one embodiment, the method calculates a room source sequence that is required to be small and have different trip durations and can reach a lower number value than a room source with an adjacent trip duration. In one embodiment, the calculation begins with the house source L conforming to the desired characteristics. In one embodiment, the computation is performed using a greedy approach. For example:
(a) The method considers the house sources U with sales prices, sorts the house sources according to the sales prices, and firstly considers the house source with the lowest sales price;
(b) Then, the method processes house sources according to the selling price sequence;
(c) During the process, the method selects the next house source l and excludes any house sources having a travel duration within the travel duration range of l in the subsequent process;
(d) Step (c) of the method is then repeated until there are no more sources to process.
The calculation will produce a certain number of k.gtoreq.1 house sources. The number k depends on the breadth of the range. For example, the range is set to 15 minutes or 1000 meters. We will rank in increasing order of travel duration, with the calculated house sources ranked as l 1 ,l 2 ,l 3 ,…,l k . For example: l (L) 1 (4001) The duration of the trip is 5 minutes, selling $ 100 tens of thousands; l (L) 2 (4002) The duration of the trip is 14 minutes, selling $ 120 ten thousand; l (L) 3 (4003) The duration of the trip is 25 minutes, selling 70 ten thousand dollars; and so on. Because of the way the house source is calculated, we know for each l i None are sold at a lower price and the duration of the journey is at l i A room source within a neighborhood of travel duration of (c). The neighborhood includes a maximum of l i Or at least l i Is provided for the duration of the stroke. The neighborhood may include both ranges. This feature of the neighborhood depends on long monotonic operation of the sales price over the duration of the journey. For example, in the above example, l 1 Is the cheapest source of any real estate with a trip duration of less than 5 minutes. In one embodiment, we calculate a series of room source sequences that are relaxed in terms of duration or number of trips. For example, we will choose at most a threshold of room sources within a range of travel durations, such as at most 5 room sources. For example, we choose at most a house source threshold with the lowest number of house sources in a neighborhood, such as at most 5 house sources. In one embodiment, the calculated house sources are arbitrarily orderedThus, l 1 ,l 2 ,l 3 ,…,l k The ascending order of the duration of the travel is not necessarily followed; for example, following a decreasing order, or a monotonic order of the number associated with the calculated house sources). It will be apparent to those of ordinary skill in the art that many other methods of calculating the house source sequence can be employed without departing from the scope and spirit of the present embodiments.
In one embodiment, the method determines a computing room source l 1 ,l 2 ,l 3 ,…,l k Is an indication of (a). While a cluster indication is described in the present disclosure, it will be apparent to those of ordinary skill in the art that 1 ,l 2 ,l 3 ,…,l k There is also a corresponding indication. For example, the indication includes a sales price and an excerpt for each calculated house source. In one embodiment, the indication includes a mathematical statistic about sales price, the statistic relating to the duration of the trip at l i All house sources within the travel duration range of (c). Some mathematical statistics examples are described in the present disclosure, such as the 10 th percentile, the 50 th percentile, and the 90 th percentile. So that the user can know the travel duration and/ i The sales price of the house source for a comparable duration of the journey. In one embodiment, the indication includes mathematical statistics of sales prices for the entire house source U. So that the user can know what the general sales concept is on the market. For example, the user may see which of the two-room apartments currently on the metropolitan area real estate market are inexpensive house sources (e.g., the 10 th percentile) (4004) and which are higher than market price house sources (e.g., the 90 th percentile) (4005).
Calculated house source l 1 ,l 2 ,l 3 ,…,l k Can be considered as "first level" house sources, i.e. they provide a minimum sales price overview based on the duration of the journey. In one embodiment, the method calculates a "second level" house source (4006). For one i, the method considers i i And its next l i+1 And their phasesDuration d of off travel i And d i+1 . The method then considers a subset U of the house sources U i With a duration of travel of d i And d i+1 Between them. The method then uses any of the methods disclosed herein to determine U i For example, a method for selecting clusters. For example, U l The indication of (a) includes a small number of geographically dispersed sources (e.g., up to 10) with a trip duration of between 5 minutes and 14 minutes and with the lowest available sales price among the sources within the trip duration. U (U) i The indication of (c) may include U i Is a mathematical statistic of (a). The set of edge cases is defined as follows: u (U) 0 Comprising a duration of a run in U less than d l All house sources of (1), U k Comprising a duration of a run in U of at least d k Is a house source; in one embodiment, the set of edge conditions is further limited to a range of travel durations. Set U i May be empty, e.g. when there is no room source meeting the desired characteristics, and its travel duration is d i And d i+1 And (5) time. Thus, the "second stage" may be non-uniform. In one embodiment, the method calculates a "second level" house source by subdividing the trip duration and the calculation sequence. In one embodiment, the method calculates a "third level" house source by further subdividing the travel duration, and so on.
In one embodiment, the method determines an overview of a graphical representation of a relationship between sales price and travel duration. In one embodiment, the graphical representation describes mathematical statistics of sales prices associated with each range of travel durations. For each range of travel durations, the method determines mathematical statistics of sales prices and generates shapes that encode the mathematical statistics, such as: points, lines, ellipses, rectangles, range bars, etc. The graphical representation takes many forms including, but not limited to: charts, histograms, pie charts, and heatmaps. For example, the chart includes: a horizontal axis corresponding to the duration of the trip, a vertical axis corresponding to the sales price, and a mathematical statistical plot of the sales price associated with the source of the room for each range of the trip duration. For example: (4010) Represents the 90 th percentile of sales prices, (4011) represents the median of sales prices, and (4012) represents the 10 th percentile of sales prices. For example, as shown in FIG. 4, in a house source where the duration of the journey is in the range of 30 minutes, the 10 th percentile of sales prices is $100 ten thousand. It will be apparent to those of ordinary skill in the art that many other methods of determining a graphical representation of the relationship between sales price and travel duration may be employed without departing from the scope and spirit of the present embodiments.
In one embodiment, the method may enable navigation and be performed by a device. The device displays an indication of the "first level" room source. The user may interact with the device. For example, the user may interact with the user interface element (4007) (such as clicking, hovering a mouse, performing a gesture, such as touching, sliding, long pressing on a touch-sensitive display screen, etc.), the device displaying an indication of the appropriate "second level" room source in response, or interact with the user interface element (4008) to hide the indication. In this way, the user sees a compact overview of the lowest sales price, allowing the user to see the overview in depth, exploring the trade-off between sales price and duration of the trip. In one embodiment, the house source is displayed in the following manner: linear (e.g., list), the user can scroll up and down through the presentation; annular (e.g., a list folded into a circle), user may rotate the display or arrange in reverse order, etc. In one embodiment, the location of the "second level" room source is presented on the map during the "second level" room source display. In one embodiment, when a "second level" room source is displayed, the locations of other room sources will be hidden from the map. In one embodiment, the user interacts with the graphical representation. For example, the user clicks on a user interface element (4013), and in response the device will display an indication of the source of the room for which the duration of the journey is within the range associated with that element. It will be apparent to those of ordinary skill in the art that many other navigation methods can be employed without departing from the scope and spirit of the present embodiments.
The method uses other ways to calculate the house source l using the sales price and the journey duration 1 ,l 2 ,l 3 ,…,l k . In one embodiment, the method uses any of the clustering methods mentioned in the present disclosure. Each l i Are selected from a cluster, for example, as the centroid of the cluster. In one embodiment, at least one additional requirement is set for the cluster, including setting a maximum value of k, e.g., 20; requirement l k The lower the better; setting l i And/or the duration of travel of (1) i+1 For example 10 minutes; setting a maximum number of selected house sources, e.g., 5, within the travel duration neighborhood; only the house sources with the shortest travel duration are required to be clustered, for example, the shortest time is 2 hours; only the lowest selling price house sources, such as 75 percentile minimum, are required to be clustered; requiring selection of l from sources having a range of travel durations l For example in the range of a minimum of 5 minutes; selecting l with approximately the lowest sales price i For example, the duration of the journey is at a value including l i Within 10% of the minimum sales price of any house source within the range of (3); etc. In one embodiment, the clustering problem with at least one additional requirement is encoded as a linear program. In one embodiment, the range of travel durations is different from the range of other travel durations. For example, the range may be narrower for a trip duration having a relatively low selling price for a house source. In practice, house Source l 1 ,l 2 ,l 3 ,…,l k Is not required to be equal. It will be apparent to those of ordinary skill in the art that many other computing house sources can be employed without departing from the scope and spirit of the present invention.
In the above, we describe one embodiment with sales price as quantity. Generally, a quantity is an entity that can be compared to other quantities to establish a sequence. In one embodiment, the number is set to a characteristic value, for example, to a school rank. In one embodiment, the quantity is described in a request, such as a request to say "prefer a high selling price", in which case the quantity is a negative number of selling prices (i.e., selling price multiplied by negative one). In one embodiment, the number is derived from the eigenvalues by a mathematical formula. For example, reflectThe number of "floor centrality" is represented by the formula (f/b-0.5) 2 Calculated, where b is the number of floors of the building and f is the number of floors of the source of the branding house in the building. In one embodiment, the device will display derived text related to the number of derivations, such as "floor centrality", not just "floor level". For example, the number reflecting similarity to the geographic orientation is determined as the absolute value of the orientation difference along the shortest arc (e.g., the number of east windows is 90 degrees relative to a user request specifying a north-facing window). In one embodiment, the quantity is derived from the values of two or more characteristics, such as dividing the sales price by the area. In one embodiment, the number is determined based on features predicted from the desired features. For example, when a user searches for an apartment with an area of about 100 square meters, the number is set to |x-100|, where x is the square meter area of the apartment. In one embodiment, predictions are made automatically based on requests and features, for example using: presetting rules; artificial intelligence software, for example, training based on past requests; etc. In one embodiment, the number is equivalent to a score. It will be apparent to those of ordinary skill in the art that many other methods of determining the amount can be employed without departing from the scope and spirit of the present invention.
In one embodiment, one element (4002) may be described as an indication of at least one room source a and the other element (4003) may be described as an indication of at least one room source B; wherein A and B are both contained in one isochrone E, and A is dissimilar to B. In one embodiment, one element (4009) may be described as an indication of at least one room source C that is also included in the same isochrone E, and a is dissimilar to C, and B is dissimilar to C. In one embodiment, isochrone E is determined to be broad. For example, given a request to find an apartment, the width is at least 160 minutes when no apartment is traveling for a duration of between 20 minutes and 180 minutes. In one embodiment, the indication of the plurality of sites comprises an overview of the plurality of sites.
It will be apparent to those of ordinary skill in the art that many other methods of determining a real estate source overview may be used in association with each real estate source and travel duration without departing from the scope and spirit of the present invention.
1.3.3 alternatives
In one case, assume that there are two schools, each school's school district including a particular residence. In this case, the child living in the house can choose to go to any one of the two school schools. Once decided to read, the child goes to the selected school and not to another (unselected) school. Thus, the selected school contributes to the overall commute time of the home. While another school does not contribute to the overall commute time, it is useful to know that it is useful to have an alternative to school for finding a residence; for example, because another school may contribute at other times.
This presents a general problem in that it is desirable to determine an indication of alternative points of interest in a manner that reduces user information overload and improves relevance. We now present a solution to this problem
We first describe an embodiment of an alternative school. Our method is for school S 0 (5003) And real estate H (5002) is operated. In one embodiment, S 0 And H are both included in the request. In one embodiment, H is included in the request, then S 0 Is the school nearest to H in terms of duration of the trip or distance. In one embodiment, S 0 Included in the request, while H is determined by the steps of the other method. In one embodiment, H and S 0 Are determined by the steps of the other methods. We determine school S using any of the methods disclosed herein 0 Duration of travel D with property H 0 (5001) (e.g., H and S 0 The round trip duration between). Then we determine an alternative school collection A whose duration of travel is relative to school S 0 Duration of travel D of (2) 0 And not much larger. For this reason we have determined an alternative school S with m.gtoreq.0 nearest to the real estate H 1 ,…,S m (5004) (5005) (5006). We determine school S i Duration of travel with real estate H. Then, toAt each i.gtoreq.1, we evaluate D i Whether t is less than or equal to t for a travel duration t (5007), the travel duration being set to D 0 A threshold is added. Examples of threshold values are 20 minutes and 2000 meters. If the assessment is successful, school S i (5004) (5005) will be included in set a. In one embodiment, if the assessment fails, school S i (5006) Will be excluded from set a. The resulting set a may be empty, for example, when each of the candidate schools is far from property H. In one embodiment, we will school S 0 Set a is included. In one embodiment, we will school S 0 Excluding set a. In one embodiment, set A includes at least two schools.
Then, we determine an indication of set a. The indication includes, but is not limited to, at least one of:
(a) Corresponding to any of the indications described in section 1.3.1;
(b) Semantics of the collection A, such as the text "average rank of candidate schools";
(c) Information about schools in collection A acquired by the acquisition module, for example: (i) school name, (ii) school type, (iii) school rank, (iv) school fare, (v) school admission criteria, or (vi) school admission probability;
(d) Mathematical statistics (5008) of information about schools contained in collection a, such as (i) number of schools, acquired by the acquisition module; (ii) highest, lowest, or average ranking of schools; (iii) a school high, lowest or average school charge; (iv) Probability of being recorded by any school, e.g. 1-pi i∈A (1-p i ) Wherein p is i Is school S i Probability of admission; (v) Expected values of rank, e.g. Σ i∈A (p i ·r i ) Wherein r is i Is school S i Is a ranking of (2); or (vi) the expected value of the school charge, e.g. Σ i∈A (p i ·u i ) Wherein u is i Is school S i Is a learning fee of (a);
(e) Duration of travel D for schools contained in collection A i For example: (i) Duration of travel D i Or (ii) a maximum, minimum, or average trip duration of the school; or (b)
(f) Combinations of any of the above, for example: (i) Weighted sums of school ranks, e.g. by travel duration D i Score D of (2) i /(∑ j∈A D j ) Weighted, or (ii) a weighted sum of school trip durations, e.g., weighted by a probability of admission p.
In one embodiment, the indication of set A includes a description of the trip. In one embodiment, the indication of set a includes an indication of a description of the non-itinerary, such as the number of schools in set a. In one embodiment, the concept of "non-travel description" is further limited to exclude prior art, as would be apparent to one of ordinary skill in the art.
In one embodiment, the indication of set A has a non-single dependency on set A. When a dependency is limited to only at most one school in set a, the dependency is referred to as a single dependency, and any other dependency is a non-single dependency. For example, the name of one school in set A is singular, but the names of the top schools in set A, including at least two schools, are non-singular. The non-single indication is very useful because it can summarize the information of a large set a. In one embodiment, the non-unity indication is a piece of statistical information that is a function of the values of each school (including at least two schools) in set A such that the statistical information has a non-zero partial derivative for each value. In one embodiment, the average ranking of at least two schools is a non-unitary indication. In one embodiment, the indication of set a comprises a single indication. In one embodiment, the indication of set a comprises a non-single indication. In one embodiment, the concept of "not singular" is further limited to exclude prior art, as would be apparent to one of ordinary skill in the art.
In one embodiment, we explicitly restrict the indication of set a such that the indication does not include any information other than the information explicitly listed by the restriction. These limitations include any indication of set A above, but with the addition of the qualifier "only", e.g
(a) Only the top-ranked school name in set a is included;
(b) Only the number of schools is included;
(c) Only the average trip duration of the school;
(d) Only the average ranking of schools is included;
(e) There are only two combinations of indicators; or (b)
(f) When k is more than or equal to 3, only k indexes are combined.
This has an advantage in that these restrictions reduce the information burden on the user and increase the relevance. For example, presenting the number of schools in set A as unique information about set A places less cognitive load on the user, while providing useful information to the user.
In one embodiment, the candidate schools S are compared prior to determining set A 1 ,…,S m Screening is performed. In one embodiment, we screen based on whether real estate H is contained within the school district of the relevant school. In one embodiment, the screening is performed upon request. For example, the request may specify: "private school only", "school limited to the furthest 20 minute distance from home", "school limited to the school with a school fee no more than $500", "school limited to the top 30", "school limited to my children who may be logged according to the school criteria", "school limited to consider my children's following features, with a logging probability of at least 80%," and so on. In one embodiment, we use a preset screening method based on a mathematical formula that uses journey duration or school eigenvalues. An example of a mathematical formula expresses the filtering specified in the request.
In one embodiment, we score schools in a similar manner to the scoring of house sources described previously. And accordingly we use the score of the school to select set a.
In one embodiment, the information acquired by the acquisition module about schools contained in collection A is used to determine a score for the real estate source. For example, the score is increased by the mathematical statistics in set a.
In one embodiment, the indication of the plurality of sites includes an indication of set a. For example, when a request specifies that the properties should be ordered in a school rank, then the indication of the properties includes an indication of the nearest school, and an indication of the alternative schools in set A. In one embodiment, any of our methods includes the computation of set A. In one embodiment, the service module responds to the request with an indication of set a.
The method related to the alternative school indication can be generalized to the alternative interest point indication. However, we need to discuss how to determine alternative points of interest S 1 ,…,S m . Whether two points of interest can be considered as alternative points depends on the specifics of the point of interest, and thus the result may be arbitrary. Thus, our method can automatically determine alternatives, which would be within the purview of one of ordinary skill in the art. For example, if the point of interest is a hospital and the request specifies "orthopedics", then this may be determined by each hospital having an orthopedics ward. In one embodiment, we use the similarity between points of interest to determine alternative points of interest S 1 ,…,S m 。
It will be apparent to those of ordinary skill in the art that many other methods of determining an indication of alternative points of interest may be employed without departing from the scope and spirit of embodiments of the present invention.
1.3.4 two stage Process
In one embodiment, the service module uses a two-stage approach: the first stage is to calculate estimated travel durations and use them to select a number of clusters; the second stage is to calculate the travel duration of the selected cluster and use the travel duration to determine the indication. In one embodiment, a two-stage approach may save resources (first stage) and limit the degradation of stroke quality (second stage). For ease of explanation, we will describe one embodiment with reference to only one commute destination (6001). However, it will be apparent to those of ordinary skill in the art that the present invention may be generalized to at least one commute destination without departing from the scope and spirit of the present invention. This embodiment operates on any set of house sources, including:
(a) Calculating estimated travel durations between each geographic location in the house source and the commute destination, implementations including
(i) Identifying a nearby one of the representatives (6003) within a threshold range of the commute destination (6001), identifying a nearby one of the representatives (6005) within a threshold range of the geographic location (6007), retrieving a pre-calculated travel duration (6004) between the two nearby representatives, and setting the pre-estimated travel duration to the retrieved travel duration, optionally using the following increase times: a duration of travel (6002) between the commute destination (6001) and its nearby representation (6003), or a duration of travel (6006) between the geographic location (6007) and its nearby representation (6005);
(ii) Identifying a nearby delegate (6003) within a threshold range of the commute destination (6001), retrieving a pre-calculated travel duration (6008) between the nearby delegate (6003) and the geographic location (6009), and setting the pre-estimated travel duration to the retrieved travel duration, optionally increasing the travel duration (6002) between the commute destination (6001) and the nearby delegate (6003);
(iii) Identifying a nearby representation (6011) within a threshold range of the geographic location (6013), retrieving a pre-calculated travel duration (6010) between the commute destination (6001) and the nearby representation (6011), and setting the pre-estimated travel duration as the retrieved travel duration, optionally enhanced with the travel duration (6012) between the geographic location (6013) and the nearby representation (6011); and
(iv) Acquiring a journey duration (6014) between the commute destination (6001) and the geographical location (6015) from the navigation service,
an example threshold is a stroke of 1000 meters or 1 minute,
in one embodiment, we identify at least one nearby representation within a threshold range of the commute destination or geographic location, and set the estimated travel duration to the minimum of any travel durations of the commute destination and geographic location through any nearby representation,
In one embodiment, we pre-compute a nearest neighbor data structure (e.g., thiessen polygon (Voronoi Cell) representing distance or travel duration), and use the nearest neighbor data structure to determine nearby representatives during request processing;
(b) Selecting one or more room source clusters using the estimated travel duration, but not exceeding a predetermined limit, the value of the predetermined limit affecting the number of clusters indicated to be determined later in step (d), the setting being based on, but not limited to, at least one of: the response needs to include the indicated number of clusters, the reduced quality of the trip due to the estimated duration of the trip in step (a), the improved quality of the trip due to the duration of the trip in step (c), the resource consumption associated with the estimated duration of the trip determined in step (a), or the resource consumption associated with the determined duration of the trip in step (c); for example, the predetermined limit is set to 1000 (e.g., by selecting a room source that matches the desired feature, clustering the selected room source, scoring the clusters, etc., the highest scoring cluster is selected);
(c) Determining a trip duration between each geographic location of the room source contained in the selected cluster and the commute destination, for example using a prior art method of calculating the shortest path or any of the methods of calculating the trip mentioned in the present disclosure; and
(d) An indication of the selected cluster is determined using the run-time duration (e.g., after updating the clusters and scores using the run-time duration; e.g., by selecting a number of clusters, such as the highest clusters, that is at most a preset portion of a predetermined limit, such as at most 20 highest clusters).
In one embodiment, the method smoothes the stroke representative of the vicinity. The smoothing process may prevent strokes near the location where the pre-computed stroke meets the enhanced stroke from assuming an unnatural shape. For example, we retrieve a pre-computed partial stroke starting from a representative position near the source position and augment the partial stroke with a stroke between the source position and a position contained in the partial stroke (which position is not necessarily the representative position). For more information on smoothing, reference is made to the prior art WO 2021222046.
In one embodiment, the method calculates the overview using a two-stage method. In one embodiment, the method calculates a "first level" house source l 1 ’,l 2 ’,l 3 ’,…,l k ' As described above, but using predictionsInstead of the duration of the stroke, and the range of use is narrow, for example only 1 minute. The value of k' will typically be large due to the narrow range. The method then determines the house source l 1 ’,l 2 ’,l 3 ’,…,l k ' duration of travel. The method then proceeds by applying to the house source l 1 ,l 2 ,l 3 ,…,l k Reshuffling and pruning, using a wider range (e.g., 15 minutes) and trip duration to calculate a "first level" house source l 1 ’,l 2 ’,l 3 ’,…,l k '. For example, the method calculates the house source l 1 ,l 2 ,l 3 ,…,l k In which the calculation is from a house source l 1 ’,l 2 ’,l 3 ’,…,l k ' Start (we describe an embodiment above in which the calculation starts from the house source L). In one embodiment, the method calculates a graphical representation using the estimated travel duration. It will be apparent to those of ordinary skill in the art that many other methods can be employed to calculate an overview using a two-stage method without departing from the scope and spirit of embodiments of the invention.
In one embodiment, we use a two-stage approach. In one embodiment, the method calculates an indication of at least two alternatives using a two-stage method, where set A is determined using an estimated duration of travel and the indication is determined using the duration of travel. In one embodiment, the method calculates the isochrone using the estimated travel duration and then calculates the indication using the travel duration. It will be apparent to those of ordinary skill in the art that many other methods may be used in our method using a two-stage method without departing from the scope and spirit of embodiments of the invention.
1.3.5 modification step
In one embodiment, the service module performs the modified steps as compared to the steps disclosed herein. For example, the service module performs the steps in other orders, may perform part of the steps simultaneously, and may perform a combination or omit the steps. In one embodiment, the grouping and clustering steps are combined into one step. For example, we extend the feature vector. We use one feature vector of the house source and add: features and values representing the duration of travel between the geographic location of the house source and at least one commute destination, while adding features and values of relevant points of interest. Feature vectors that are expanded in this way will be clustered. In one embodiment, the clusters are determined using any of the clustering methods disclosed herein. In one embodiment, the clusters meet additional requirements. For example, we limit any clusters to span a range of up to a few minutes, e.g., up to 15 minutes, along the axis of the added feature. In one embodiment, the pre-computed clusters are used to accelerate the clustering during request processing, e.g., by a clustering algorithm, performed starting with the pre-computed clusters. In one embodiment, the step of identifying a cluster does not use similarity. In one embodiment, this step determines clusters of at least one location. In one embodiment, this step uses any information contained in the disclosed request to identify at least one location, such as: screening for limitations or desired characteristics. It will be apparent to those of ordinary skill in the art that many other methods of performing the modified steps can be used without departing from the scope and spirit of the embodiments of the invention.
General case 2
We use the term "travel" in a broad sense, consistent with the interpretation of the word by one of ordinary skill in the art. The meaning of the term includes moving objects or data. The description of the trip is anything that one of ordinary skill in the art would consider. The following are examples of some descriptions of travel: (1) "he, you need to walk north one block, then turn left slightly", and (2) "dollar 5". The stroke length is a value that one of ordinary skill in the art can associate with a stroke, such as: the monetary cost of the journey; metric distance; fuel consumption; specific features or attributes of the description of the journey, for example: number of transfers or walking distance. For another example, when we refer to a run length representing time, we can use the term run duration. In one embodiment, the duration of the stroke is derived from any end of the stroke, including: real estate, commute destination, or any feature or value thereof. For example, the run length is derived using a weighted sum of two values: (1) metric distance from company exit to building entrance; (2) characteristic values representing real estate floors in a building. In one embodiment, the run length uses the request. For example, the request includes any transition between two values. For example, the stroke length is the fuel consumption amount multiplied by the conversion rate from the unit fuel to the monetary amount, thereby converting the optimization target of using the fuel consumption amount into the optimization target of using the monetary cost. In one embodiment, the run length uses any pre-set scaling factor. In one embodiment, the run length uses two or more optimization objectives that are combined into one optimization objective, for example using an arbitrary mathematical formula, such as a weighted sum. In one embodiment, the optimization objectives are combined into a multi-objective optimization search based on multi-dimensional costs. For example, a method may search for a real estate whose trip duration is the shortest, subject to trip monetary costs and to screening constraints. The duration of the stroke itself is a description of the stroke. Description of the travel: the length of any stroke may not be included, the extent of the stroke may be included only, and some other data may be included. The present invention discloses a description of how to calculate a trip, for example using any of the methods of calculating a trip described in the present disclosure, or any of the prior art of calculating a trip mentioned in the present disclosure, for example using the Dijkstra algorithm.
We use the term "traffic system" in a broad sense, consistent with the interpretation of this term by one of ordinary skill in the art. Some embodiments include: road and automotive systems; a public transportation system consisting of buses and subways; a walking path system; airports, airplanes and waterways; or vessels and marine channels. The traffic system does not require actual moving objects. The method disclosed by the invention only needs to be capable of determining the description of the journey among the elements of the traffic system. Thus, a traffic system that moves data is one example of a traffic system. For example, a computer network consists of the following transport elements: wires/lines (similar to roads) and hubs/switches (similar to stops/turns). Any combination of traffic systems that allow transfer between them is a traffic system. It will be apparent to those of ordinary skill in the art that many other examples of traffic systems may be employed without departing from the scope and spirit of embodiments of the present invention.
In one embodiment, our invention relates to a search or comparative embodiment other than real estate. In one embodiment, our method utilizes the similarity between job recruitment information and the duration of travel between the work location and the home location to present the recruitment information. For examples of searches or comparisons, please refer to prior art WO 2021222046.
In general, one method of the present disclosure uses any location (in the previous section, one location is referred to as real estate) and any location (in the previous section, one location is referred to as commute destination) contained in a transportation system, and uses at least one description of a trip (in the previous section, the duration of the trip is summarized as a description of the trip) between at least one location and at least one location to determine an indication of the at least one location. The location is an arbitrary position. It may be any real estate such as apartments, rentals, garden houses, pastures, hotels, etc.; it may also be a work site, restaurant, store, etc. The place is also an arbitrary position. It includes schools, grandparents 'families, golf courses on weekends, favorite restaurants, doctor's offices, places to worship, and the like. It may also be a place where a person resides. In one embodiment, a point of interest is interpreted as a place. In one embodiment, a point of interest is interpreted as a place.
One embodiment is a method of searching or comparing at least one location using at least one description of a journey within a transit system between the at least one location and the at least one location, the method comprising: (a) receiving a request comprising at least one place; and (b) responding to the request using at least one search or comparison result obtained from the description of the trip. In one embodiment, the result of the search or comparison is an embodiment of an indication of at least one location.
In one embodiment, the methods disclosed herein perform variations of the functions or steps described previously. In one embodiment, certain functions or steps may be performed in other sequences, partially concurrently, and possibly in combination or omission. For example, one method executes the service module, but does not execute the acquisition or indexing module. In another example, the indexing module (1003) does not generate any inverted index (1005), or does not generate any cluster (1006). In another example, the request does not contain a description of at least one commute destination. In one embodiment, a method performs clustering or scoring without using a duration of travel between a real estate and at least one commute destination. In one embodiment, one of the following is used, but not both, to determine multiple house sources: (i) Similarity between house sources, or (ii) travel duration between the geographic location of the house source and at least one commute destination. It will be apparent to those of ordinary skill in the art that many other methods of performing the functions or step variants can be employed without departing from the scope and spirit of the embodiments of the present invention.
Aspects of the invention may take the form of a hardware embodiment, a software embodiment or a combination of both. The steps of the invention, such as any of the blocks of the flowcharts, may be performed out of order, partially concurrently, or provided from cache, depending on the functionality or optimization. Aspects may take the form of a sequential system or a parallel/distributed system, with each component embodying certain aspects and possibly repeated with other components; the components may communicate with each other, for example, using any type of network. The present invention is not related to any particular programming language. The computer program containing instructions for performing the steps of aspects of the present invention may be written in any programming language, such as C++, java, or JavaScript. Any program may be executed on any hardware platform, such as a Central Processing Unit (CPU) or Graphics Processor (GPU) and associated memory or storage devices. Programs may perform various aspects of the present invention within one or more devices, including but not limited to: a smart phone running an android or iOS operating system, or a web browser, such as a Firefox (Firefox) browser, a Chrom browser e, an IE browser, or a Safari browser.
3 method
Embodiments of the invention include the following methods:
1. A method of determining an indication of a plurality of locations within a traffic system using a trip length and a similarity, the method characterized by:
(a) Receiving a request included in at least one place within the traffic system;
(b) Determining at least two isochrone locations contained in the plurality of locations, wherein a travel length between each isochrone location and the at least one location within the transportation system is contained within a range;
(c) The determination is made using one of the following steps
i. Determining a plurality of similar places included in the at least two isochronal places, and determining the indication of the plurality of similar places; or alternatively
Selecting at least one first location that is dissimilar from at least one second location and that is contained in both the at least two isochronal locations, and determining an indication of the at least one first location and the at least one second location;
and
(d) Responding to the request with the indication.
2. A method of determining an overview of a plurality of locations within a traffic system using a trip length and a quantity, the method characterized by:
(a) Receiving a request included in at least one place within the traffic system;
(b) The calculation is included in theIn a sequence of two or more places, wherein,
i. for a first location and a second location included in the sequence, the first location and the at least one location are separated by at least a range between "a length of travel within the transit system" and the second location and the at least one location between "a length of travel within the transit system", and
so long as the travel length between the fourth location and the at least one location is within the traffic system
Including the vicinity of the "length of travel within the transit system" between the third location and the at least one location,
then
The number associated with the third location included in the sequence is at most equal to the number associated with the fourth location included in the plurality of locations;
(c) Determining the overview comprising the sequence indication; and
(d) Responding to the request with the overview.
3. A method of determining an indication of at least two alternatives among a plurality of points of interest within a transportation system, the method characterized by:
(a) Receiving a request including a location within the transportation system;
(b) Determining the at least two alternatives, wherein,
The length of travel between each alternative and the location within the traffic system is within a threshold of a shortest travel;
(c) Determining the indication of the at least two alternatives, wherein the indication is a non-unitary and non-stroked description; and
(d) Responding to the request with the indication.
4. A method of determining an indication of at least two locations within a traffic system using a length of travel and a length of travel estimated, the method characterized by:
(a) Receiving a request included in at least one place within the traffic system;
(b) At least two estimated stroke lengths are determined, wherein,
the at least two estimated travel lengths include an estimated travel length within the traffic system between the at least one location and each of the at least two locations;
(c) Selecting one or more of the at least two locations using the at least two estimated travel lengths, wherein the number of the one or more locations is at most a predetermined limit;
(d) Determining at least one travel length, the at least one travel length comprising a travel length within the traffic system between the "each of the one or more locations" and the "at least one location";
(e) Determining the indication of the one or more locations using the at least one run length; and
(f) Responding to the request with the indication.
4 computer system and apparatus
As shown in FIG. 1, one embodiment of the present invention is a computer system. The computer system has a hardware embodiment, a software embodiment, or a combination of both. The computer system includes at least one processor, such as a CPU or GPU. The computer system includes a non-transitory computer readable storage medium storing one or more programs for execution by at least one processor. Embodiments of non-volatile computer-readable storage media are well known in the art and need not be described herein. The one or more programs include instructions for execution by the at least one processor to perform at least one of the steps of the methods disclosed herein. In one embodiment, the instructions may be expressed in any programming language, such as C++, java, or JavaScript. Each method produces a computer system. Any such computer system can be considered to be a general purpose computer specifically programmed to perform the particular methods described in the present disclosure. Thus, in practice, the computer system is a special purpose computer programmed with instructions encoding the software (one or more programs) of the method to perform the specified steps of the method. It will be apparent to those of ordinary skill in the art that many other embodiments of the computer system may be employed without departing from the scope and spirit of the embodiments of the present invention.
As shown in fig. 3, 4 and 5, one embodiment of the present invention is an apparatus, which may also be referred to as a device. It will be apparent to those of ordinary skill in the art that modifications may be made to the apparatus as shown in the accompanying drawings without departing from the scope and spirit of the embodiments of the invention; for example, components may be rearranged, resized, color changed, shape changed, added or removed, and the like. The device has an embodiment of entity form, such as a smart phone application program or a web page. The device receives a user's request via a "receiver" (e.g., a user interface of a smartphone application), e.g., the user may enter the location of the workplace in a search box, click on a map displayed by the smartphone application to specify the location of the workplace, provide GPS readings encoding the location of the workplace via a dialogue with speech recognition to describe the location of the workplace, etc. In one embodiment, the receiver receives any information contained in the request of the present invention. The device then uses the methods disclosed herein to generate an indication. In one embodiment, the generation of the instructions is accomplished by executing an appropriate program or programs on an appropriate at least one processor. The device then responds to the user by means of a "transmitter", for example: as shown in fig. 3, a display screen (3005) in a smartphone application, a speech synthesizer (e.g., by talking to the user), AR glasses in glasses mounted on the user's head, a 3D projector of the warrior, and so on. It will be apparent to those skilled in the art that the present invention is not limited to a device, nor to a receiver, nor to a transmitter. It will be apparent to those of ordinary skill in the art that many more embodiments of the invention exist without departing from the scope and spirit of the embodiments of the invention.
5 ending language
It will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the scope and spirit of the embodiments of the invention. In addition, the description of the present invention may be modified in accordance with the specific circumstances without departing from the scope and spirit of the embodiments of the present invention. Thus, while embodiments of the invention have been illustrated and described, such description is not intended to be limiting, and the invention should not be limited to only those embodiments. Rather, all embodiments that come within the scope of the following claims are intended to be embraced therein.
The method of any claim does not include a "psychological process", i.e. any step of the method of any claim is not performed in the human brain. The method of each claim is automated and examples of automation are described in section 4. The scope of the method of each claim does not include any embodiments in which the particular jurisdiction in which the patent application is filed at the PCT national/regional stage does not comply with the patent application. For example, the application filed in Canadian implicitly indicates that the method of each claim is limited to embodiments that meet the requirements of the application in Canadian. Each particular jurisdiction excludes embodiments that are specific to that jurisdiction, i.e., different jurisdictions may exclude different embodiments.
In one embodiment, any of the claimed methods are implemented on or by a computer system and are used for purposes of implementation on a device, such as searching or comparing or determining an indication. The invention is described in section 4 as an example. It will be apparent to one of ordinary skill in the art that in one embodiment, any claimed method is limited to embodiments within the scope of "manufacturing modes" as defined in monopoly regulations for use in New Zealand. It will be apparent to one of ordinary skill in the art that in one embodiment, any claimed method is limited to embodiments within the scope of "technical" as defined in European patent convention.
Any prior art cited in this disclosure is considered a common knowledge in the art; any person of ordinary skill in the art will have this knowledge.
The premise basis in the claims is sometimes identified by boxes for querying: as defined in claimCan be used as +.>
We list the glossary of terms appearing in the claims, as well as reference examples in the specification. These references are not exhaustive; other references also exist. The order of the phrases in the table are intended to follow the sequence in which the terms appear in the claims.
/>
/>
/>
Claims (4)
1. Determination using run length and similarityInner->Is->Is->The saidIs characterized in that:
(a) The receiving is included in theInner->Is->
(b) Determining inclusion in theIs->Wherein,,
each isochrone site and theBetween said->The stroke length in is contained within a range;
(c) The determination is made using one of the following steps
i. Determining inclusion in theIs->And determining theIs>Or alternatively
Selection ofAnd->Dissimilar and saidIs in contact with the->Are all included in said-> And determining said +.>And said->An indication of (2); and
(d) With the saidResponsive to said->。
2. Using stroke length and quantity determinationInner->Is->Is->The saidIs characterized in that:
(a) The receiving is included in theInner->Is->
(b) The calculation is included in theIs +.>Wherein,,
i. for the saidIs comprised of->And->
The saidAnd said->Between said->Inner stroke length and
the saidAnd said->Between said->At least one range of travel lengths, and
provided that
And said->Between said->The stroke length is contained in
And said->Between said->In the vicinity of the stroke length in the inner zone,
then
Is included in theSaid- >The number associated is at most equal to the number contained in said +.>Said->The associated number;
(c) Determining to contain theSaid->And
(d) With the saidResponsive to said->
3. Determining oneInside multiple points of interest +.>Is->Is-> Said->Is characterized in that:
(a) The receiving is included in theInner one->Is->
(b) Determining the saidWherein,,
each alternative and theBetween said->The inner stroke length is within a threshold of a shortest stroke;
(c) Determining the saidIs>Wherein said->Is a non-singular and non-stroking description; and
(d) With the saidResponsive to said->
4. Determination using a length of estimated travel and a length of travelInner->Is->Is->Said->Is characterized in that:
(a) The receiving is included in theInner->Is->
(b) Determination ofWherein,,
the saidComprises said->And said->Between each site comprised in said +.>A predicted stroke length within;
(c) Using the saidSelecting said->Is comprised of-> Wherein said->Up to a predetermined limit;
(d) Determination ofSaid->Comprises said-> Each site comprised in (c) and said +.>Between said->Inner stroke length;
(e) Using the saidDetermining said->Is >And
(f) With the saidResponsive to said->
Applications Claiming Priority (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US63/130,693 | 2020-12-27 | ||
| US63/168,279 | 2021-03-31 | ||
| US202163237535P | 2021-08-27 | 2021-08-27 | |
| US63/237,535 | 2021-08-27 | ||
| PCT/US2021/065165 WO2022140704A1 (en) | 2020-12-27 | 2021-12-24 | A method for presenting sites using their similarity and travel duration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116997923A true CN116997923A (en) | 2023-11-03 |
Family
ID=88521813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202180094592.6A Pending CN116997923A (en) | 2020-12-27 | 2021-12-24 | Method for displaying places by using place similarity and travel duration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116997923A (en) |
-
2021
- 2021-12-24 CN CN202180094592.6A patent/CN116997923A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | A trajectory clustering approach based on decision graph and data field for detecting hotspots | |
| Hu et al. | Automated urban land-use classification with remote sensing | |
| JP7264357B2 (en) | Method and Apparatus for Searching or Comparing Sites Using Routes or Route Distances Between Sites and Locations in a Transportation System | |
| Krueger et al. | Visual analysis of movement behavior using web data for context enrichment | |
| Kounadi et al. | Population at risk: Using areal interpolation and Twitter messages to create population models for burglaries and robberies | |
| Aydinoglu et al. | Implementing a mass valuation application on interoperable land valuation data model designed as an extension of the national GDI | |
| Xu et al. | Application of a graph convolutional network with visual and semantic features to classify urban scenes | |
| Fisher-Gewirtzman | Integrating ‘weighted views’ to quantitative 3D visibility analysis as a predictive tool for perception of space | |
| Brelsford et al. | Optimal reblocking as a practical tool for neighborhood development | |
| Deng et al. | Identify urban building functions with multisource data: A case study in Guangzhou, China | |
| Bi et al. | How built environment impacts online car-hailing ridership | |
| JP2024038371A (en) | Methods for indicating sites using similarity and journey duration | |
| Lin et al. | Extracting urban landmarks from geographical datasets using a random forests classifier | |
| Zhou et al. | Familiarity-dependent computational modelling of indoor landmark selection for route communication: a ranking approach | |
| Ortega-Momtequín et al. | Ranking residential locations based on neighborhood sustainability and family profile | |
| Cohen et al. | Machine‐learning prediction models for pedestrian traffic flow levels: Towards optimizing walking routes for blind pedestrians | |
| JP7144818B2 (en) | Method and Apparatus for Searching or Comparing Sites Using Routes or Route Distances Between Sites and Locations in a Transportation System | |
| Niu et al. | Integrating multiple data to identify building functions in China’s urban villages | |
| Wang et al. | What makes a place special? Research on the locality of cities in the Yellow River and Rhine River Basins based on Street View Images | |
| Huang | Simulating individual work trips for transit-facilitated accessibility study | |
| Zhang | Developing a comprehensive framework for property valuation using 3D and remote sensing techniques in China | |
| Yu et al. | Exploring Taxi Demand Distribution of Comprehensive Land-Use Patterns Using Online Car-Hailing Data and Points of Interest in Chengdu, China | |
| Thomason et al. | Context trees: Augmenting geospatial trajectories with context | |
| CN116997923A (en) | Method for displaying places by using place similarity and travel duration | |
| Sun et al. | Analysis of spatial form and structure of commercial pedestrian blocks based on Isovist and big data |
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 |