Method for predicting extended time period and related electronic device

Abstract

A method performed by an electronic device for predicting an extended period of time is disclosed. The method includes obtaining a shipping reservation request associated with a user. The method includes obtaining historical data associated with the user based on the subscription request. The method includes predicting an extended period of time based on the historical data and the subscription request (e.g., by applying a prediction model to the historical data). The method includes providing a subscription response to the user based on the predicted extended period of time.

Description

[0001] This disclosure relates to the field of transportation and shipping. This disclosure relates to a method and related electronic device for predicting extended time periods. Background Technology

[0002] Cargo shipment involves the arrival of goods at a terminal (e.g., a port) and / or subsequent departure from the terminal. Shipment of goods via a terminal (e.g., a port) involves managing demurrage and / or retention. Users may find it difficult to effectively manage demurrage and / or retention of shipments. Summary of the Invention

[0003] Users who book shipments can choose to extend the demurrage and / or delay (D&D) period for their goods. However, booking agents often find it difficult to determine and propose suitable extension periods for users to choose when booking shipments.

[0004] Therefore, there is a need for an electronic device and method for predicting extended time periods that mitigates, alleviates, or resolves existing drawbacks and allows for more accurate, robust, and / or user-centric (e.g., personalized) prediction of extended time periods provided to users.

[0005] A method for predicting an extended time period, performed by an electronic device, is disclosed. The method includes acquiring a shipping booking request associated with a user. The method includes acquiring historical data associated with the user based on the booking request. The method includes predicting the extended time period based on the historical data and the booking request (e.g., by applying a predictive model to the historical data). The method includes providing a booking response to the user based on the predicted extended time period.

[0006] An electronic device is disclosed, comprising a memory circuit system, a processor circuit system, and an interface circuit, wherein the electronic device is configured to perform any of the methods disclosed herein.

[0007] A computer-readable storage medium is disclosed for storing one or more programs, the one or more programs including instructions that, when executed by an electronic device, cause the electronic device to perform any of the methods disclosed herein.

[0008] The advantage of this disclosure is that the disclosed electronic device and method provide users with predicted extension periods by considering historical data associated with them. In other words, the disclosed predicted extension periods are personalized based on the user and the shipment, thus being more optimized. This can lead to more efficient disposal of resources (such as port equipment, transport, and machinery). Furthermore, the disclosed method and electronic device benefit from the characterization of historical data patterns to provide users with more accurate booking response time extensions. This can allow users to minimize and / or avoid D&D penalties (e.g., fees). Attached Figure Description

[0009] The above and other features and advantages of this disclosure will be readily apparent to those skilled in the art from the following detailed description of exemplary embodiments with reference to the accompanying drawings, in which: Figure 1 This is a schematic diagram illustrating an example process by which an example electronic device according to this disclosure performs an example method for predicting an extended time period. Figure 2A -B is a representation illustrating an example decision tree according to this disclosure. Figure 3A -B illustrates a flowchart of an exemplary method for predicting an extended time period performed by an electronic device according to the present disclosure. Figure 4 It is a user interface example representation based on the example output of this disclosure, and Figure 5 This is a block diagram illustrating an exemplary electronic device according to the present disclosure. Detailed Implementation

[0010] Various exemplary embodiments and details are described below with reference to the accompanying drawings, where applicable. It should be noted that the drawings may be drawn to scale or not, and elements with similar structure or function are indicated by the same reference numerals in all the drawings. It should also be noted that the drawings are intended only to facilitate the description of embodiments. The drawings are not intended as an exhaustive description of this disclosure or a limitation on the scope of this disclosure. Furthermore, the illustrated embodiments need not possess all the aspects or advantages shown. Aspects or advantages described in connection with a particular embodiment are not necessarily limited to that embodiment and can be practiced in any other embodiment, even if not so shown or explicitly described.

[0011] For clarity, the accompanying drawings are schematic and simplified, and only details that aid in understanding this disclosure are shown, while other details are omitted. Throughout, the same reference numerals are used for the same or corresponding parts.

[0012] A shipment can be considered as the shipment (e.g., transportation, such as transport from a first location to a second location) of one or more items (e.g., goods). For example, a shipment can be considered as the shipment of a container (e.g., a container containing items). A shipment may involve one or more modes of transport, such as sea, land, and / or air transport.

[0013] Demurrage can be considered, for example, the time from the arrival of a container at a terminal and / or port to the collection of the container from the terminal and / or port for transport to the recipient and / or destination (e.g., unloading at least a portion of the container). Detention can be considered, for example, the time from the collection of a container from a terminal and / or port (e.g., unloading) until the container is returned to the terminal and / or port (e.g., after unloading). For example, the returned container may contain fewer items than before it was transported from the terminal and / or port to the recipient. In some examples, the returned container may be empty. In this document, demurrage and / or detention may be referred to as D&D.

[0014] A booking request can be viewed as a shipment booking request. For example, a user can submit a booking request via a booking platform. In some examples, the user can be a booking agent. For example, a booking request includes information associated with the booking. In some examples, a booking request can be viewed as a shipment request for one or more items (e.g., goods). For example, a booking request is based on user input provided by the user. In one or more examples, a booking request includes a user code, a country code, a commodity code, and a contract code. The user code, country code, commodity code, and / or contract code can be viewed as a shipping attribute (e.g., a characteristic). A user code, for example, indicates a user or is associated with a user. For example, a user code can be viewed as a user identifier associated with a user, such as an identifier uniquely associated with a user. A user identifier can, for example, include codes indicating user identity (e.g., including one or more letters and / or numbers, such as alphanumeric codes). A country code, for example, indicates a country. For example, a country code includes one or more letters and / or numbers indicating a country, such as alphanumeric codes. A commodity code, for example, indicates a commodity. For example, a commodity code can be viewed as a commodity number (e.g., a Harmonized System (HS) code). A commodity code can, for example, include codes (e.g., alphanumeric codes, numeric codes, strings, etc.). A contract code, for example, indicates the contract associated with a shipment. In other words, a contract code can indicate the contract (e.g., agreement) performed on which a shipment (e.g., a historical shipment) associated with a user is based. A contract code may include, for example, codes (e.g., alphanumeric codes, numeric codes, strings, etc.). A contract code may be, for example, a contract number.

[0015] For example, the user is an electronic device (such as an electronic device communicatively coupled to the electronic device disclosed herein, e.g., Figure 5 The user of the electronic device 300).

[0016] Users can book shipments when (e.g., by referring to...) Figure 5When an electronic device 300 in the system provides a booking request, it selects an extended time period. For example, a user can select an extended time period to allow more time for D&D, including loading and / or unloading of containers. For example, an extended time period (e.g., free extended time) can be considered as a time period that a user can select for a given shipment. An extended time period can be considered as an extension of the D&D deadline, such as the last day of D&D retention before billing begins. In other words, an extended time period can be considered as an extension of the D&D period (e.g., free time), such that the duration of D&D may be longer before (e.g., for the user) charges are incurred. In other words, an extended time period can be considered, for example, the number of free D&D days.

[0017] Historical data associated with a user can be obtained based on a booking request. Historical data can be viewed, for example, as previous (e.g., past) data associated with the user (e.g., one or more shipments by the user). In some examples, historical data can be viewed as user-specific historical data. For example, historical data can be viewed as user-specific historical D&D data. In some examples, predictive models (such as machine learning predictive models) are used to process user-related historical data to predict (e.g., derive) the optimal extended timeframe for the user and shipment.

[0018] In other words, the extended time period can be viewed as an extended time slot, such as a D&D time slot. In some examples, the predicted extended time period can be viewed as a personalized predicted extended time period. For example, the predicted extended time period can be provided to the user (e.g., end user, booking agent, customer) when the user books a shipment (e.g., in the booking response). For example, the predicted extended time period can be provided to the user of an electronic device (e.g., via...). Figure 5 Interface 303, for example, via Figure 4 (The user interface shown). In this disclosure, the term "extended time period" may be used interchangeably with the term "extended time parameter".

[0019] Users can mitigate the risk of unforeseen (potentially costly) delays caused by demurrage and detention (D&D) when a stage of the shipment process exceeds the agreed-upon timeframe (e.g., the timeframe agreed upon by the consignee and shipper before shipment). However, consignees may not be able to mitigate the risk of unforeseen time extensions (e.g., D&D). Therefore, when submitting a booking request, users may select an inappropriate (e.g., too short) extension period, leading to one or more of the following: the carrier (e.g., a container) remains at the port longer than agreed-upon time, or the carrier fails to return to the port within the agreed-upon timeframe (e.g., after being collected). These potential events can respectively result in demurrage or detention charges (e.g., fees charged by terminal management authorities). This can lead to disputes, lost revenue, and ultimately, user dissatisfaction.

[0020] This disclosure proposes a method to predict extension periods, enabling booking responses to be provided based on predicted extension periods determined for users, thereby allowing users to choose more suitable extension periods. In other words, this method aims to improve the accuracy of predicted D&D extension periods. This can reduce the risk of D&D-related costs and associated user dissatisfaction.

[0021] Figure 1 This is an illustrative representation of an example electronic device according to this disclosure (such as electronic device 300 herein). Figure 5 The electronic device 300 illustrates a diagram of an example process by which the disclosed method for predicting an extended time period is executed. For example, Figure 1 This can be viewed as illustrating an end-to-end booking process involving electronic device 300 (e.g., a time extension predictor).

[0022] Figure 1 An example booking system 1 is shown. Booking system 1 can be, for example, a self-service instant booking system (SSIB). Booking system 1 includes a booking engine 6, and an electronic device 300 disclosed herein (such as...). Figure 5 Electronic devices 300 and database 16.

[0023] Figure 1User 2 is illustrated, such as user 2 in booking system 1. As an example, user 2 can book (e.g., request) a shipment via booking engine 6 (e.g., a booking portal). For example, booking a shipment includes booking (e.g., requesting) an extended period (e.g., D&D free time days). User 2 can provide a booking request 4 to booking engine 6. Booking engine 6 can provide a user interface through which user 2 can input and / or receive information from electronic device 300. In some examples, user 2 can provide booking request 4 directly to electronic device 300. Booking engine 6 can provide (e.g., based on booking request 4) a request 8 for a predicted extended period parameter.

[0024] Reservation request 4 may include, for example, a user code, a country code, a product code, and / or a contract code. For instance, reservation request 4 may include information indicating user 2, such as a user code that can be considered a user identifier. For example, reservation request 4 may include a user code indicating user 2 (e.g., a user identifier).

[0025] Electronic device 300 includes a processor circuitry 302 configured to perform the disclosed techniques. For example, database 16 may be configured to store historical data associated with user 2. Electronic device 300 may provide database 16 with a request 14 for historical data associated with user 2. Upon receiving request 14 for historical data associated with user 2, database 16 may provide historical data 18 associated with user 2. In other words, electronic device 300 may be configured, for example, to retrieve (e.g., via a booking portal) historical data 18 associated with a user based on a booking request (such as based on one or more elements in the booking request, such as the user code provided in the booking request) (e.g., from database 16).

[0026] For example, database 16 is an electronic data storage device. In some examples, database 16 is a remote data storage portion of an external electronic device (e.g., an external server) (such as an electronic device different from electronic device 300). In some examples, electronic device 300 includes database 16. For example, the storage circuitry of electronic device 300 (such as...) Figure 5 The storage circuit system 301 in the middle may include database 16.

[0027] For example, electronic device 300 is configured to train and / or run a predictive model (e.g., via processor circuitry 302, such as...). Figure 5The processor circuitry system 302). In some examples, the electronic device 300 includes a prediction engine 50 configured to execute the prediction models disclosed herein. For example, the electronic device 300 and / or the prediction engine 50 are configured to predict the extended time period based on historical data 18 and the booking request 4 (e.g., by applying the prediction model 50 to the historical data 18). In some examples, the predictor 50 includes a training system 10 and / or a classifier 12. For example, the prediction engine 50 is configured to train a prediction model (such as a machine learning prediction model, such as a decision tree classifier) ​​based on historical data, such as historical data associated with user 2. In other words, the electronic device 300 may be configured to classify the historical data 18 using the classifier 12 to provide the predicted extended time period 20 (e.g., by applying the prediction model (such as a decision tree classifier) ​​to the historical data 18 associated with user 2). In other words, the electronic device 300 is configured, for example, to predict the user's extended time period (e.g., a free extended time period).

[0028] In some examples, electronic device 300 is configured to categorize the predicted output into one or more time slots. The one or more time slots can be considered configurable time slots. In other words, the predicted extended time period may include one or more time slots for the user to select. For example, one or more time slots provided in the booking response are configured (e.g., customized) for the user to allow the user to select a time slot from the provided personalized time slots. For example, one or more time slots may include: Time Slot 1: Day 1 to Day 5, Time Slot 2: Day 6 to Day 10, and / or Time Slot 3: Day 11 to Day 15. For example, the extended time parameter may include one or more time slots (e.g., Time Slot 1, Time Slot 2, Time Slot 3, etc.).

[0029] In one or more examples, electronic device 300 is configured to predict extended time periods, such as one or more time slots, based on historical data 18 associated with a user. In other words, predicting extended time periods can be viewed as predicting time slot information based on historical data (e.g., historical records) associated with the user.

[0030] Electronic device 300 can be configured (such as via...) Figure 5The interface 303 shown provides a booking response 20 to the booking engine 6 or user 2 based on the predicted extended time period 20. The booking engine may provide a booking response 22 to user 2. Booking responses 20 and 22 may include, for example, the predicted extended time period 20. For example, booking responses 20 and 22 may be considered as extended time responses. In other words, the predicted free time period may be provided (e.g., transmitted and / or displayed) to the user via the booking engine (e.g., running a booking portal, such as SSIB). In some examples, the predicted extended time parameter may be considered as an extended time period adjusted for the user based on the user's historical data.

[0031] In some examples, electronic device 300 is configured to predict extended time periods, such as the number of extended D&D days, based on historical data 18. In some examples, the predicted extended time period includes analyzing historical data to train a predictive model and running the predictive model 18 on the historical data. For example, booking responses 20, 22 include time slots representing the predicted extended time periods. Booking response 22 can be viewed as enabling user 2 to select a personalized time slot (e.g., a "best" time slot) for D&D. For example, user 2 can select an extended time period based on the booking response.

[0032] Figure 2A This illustrates a representation of an example decision tree 60 according to this disclosure. Decision tree 60 is associated with a decision tree classifier. In one or more example methods, the method includes applying a predictive model to historical data associated with a user to determine the extended time period disclosed herein. In some examples, the predictive model is a decision tree classifier configured to take historical data associated with a user as input and output the predicted extended time period.

[0033] For example, historical data may include one or more of the following: one or more attributes, one or more tags associated with each attribute, a count for each attribute, and one or more time extension parameters. For example, one or more attributes (e.g., A and B) may be considered as the type of shipment feature. For example, one or more attributes may be the country (e.g., destination country or departure country), contract code, commodity code, user code, etc. In some examples, one or more attributes may be considered as decision variables (e.g., decision variables in a decision tree). For example, one or more attributes may be considered as decision parameters. For example, one or more attributes may be associated with one or more tags.

[0034] One or more labels associated with each attribute (e.g., shown by A and B) (e.g., shown by A1 to A7 and B1 to B2 in decision tree 60) can be considered as values ​​of one or more attributes of the shipment. For example, when the attribute is country, the label could be "India, United Kingdom, Denmark, United States, or Japan," etc. For example, when the attribute is contract code, the label could be the contract code of the shipment (e.g., "57"), such as the contract code of a historical shipment associated with the user.

[0035] The count of each tag associated with an attribute can be viewed as the number of times a shipment (such as a historical shipment) associated with a user is associated with a given tag of the attribute. For example, the count is associated with a tag (e.g., A3). For example, the count of each attribute can be viewed as the number of shipments associated with a given tag of one or more attributes (e.g., the amount of historical data associated with a user).

[0036] Historical data includes extended time parameters (e.g., 1-3). One or more extended time parameters can be considered as parameters indicating an extended period of time. For example, extended time parameters can include values ​​(e.g., decimals, integers, percentages, etc.).

[0037] In some examples, the method includes, for instance, determining one or more attributes, one or more labels associated with each attribute, a count for each attribute, and / or one or more extension time parameters by applying a predictive model to the initial historical data associated with the user. The initial historical data can be considered unclassified booking data, thus excluding attributes, counts, labels, actual previous extension time periods, etc.

[0038] Figure 2A Decision tree 60 is shown. For example, decision tree 60 is trained and / or generated based on historical data associated with users. Figure 2AThe decision tree shown can be viewed as a general decision tree. In some examples, decision tree 60 can be viewed as illustrating a decision tree classifier. In some examples, decision tree 60 can be viewed as illustrating a tree-structured classifier. Decision tree 60 includes nodes. Decision tree 60 includes internal nodes and leaf nodes. Node A and node B can be viewed as internal nodes of decision tree 60. For example, internal nodes of decision tree 60 can be viewed as nodes with one or more child nodes (such as those associated with one or more child nodes). For example, an internal node can be a root node, such as node A. Root nodes and internal nodes can be viewed as representing attributes provided in historical data (e.g., country, contract code, etc.). Leaf nodes can be viewed as nodes in decision tree 60 that have no child nodes (such as those unrelated to child nodes) (e.g., nodes 1 to 3). For example, leaf nodes represent the outcome of a predictive model, such as providing the category of the predicted extended time period. In some examples, leaf nodes can indicate the predicted extended time period. For example, nodes 1, 2, and 3 indicate time slots 1, 2, and 3, respectively. For example, time slot 1: day 1 to day 5, time slot 2: day 6 to day 10, and time slot 3: day 11 to day 15.

[0039] For example, a decision tree classifier includes one or more branches (e.g., one or more edges). Each branch can be viewed as connecting two nodes in the decision tree. For example, a branch represents a decision rule. A decision rule can be viewed as a set of conditions for classifying historical data (e.g., historical records). In other words, each branch represents the result of a decision rule between a parent node and its child nodes, and each leaf node represents a class label (e.g., a decision made after calculating all the attributes of the nodes that form the decision tree). In other words, a path from the root to the leaf can represent a classification rule. For example, a decision tree (e.g., a classification tree) is a tree in which each internal (non-leaf) node is labeled with an attribute (e.g., an input feature). For example, a branch emanating from a node labeled with an attribute is labeled with each of the possible values ​​of the target attribute, or a branch leads to a lower-level decision node on a different attribute. For example, each leaf node of the tree is labeled with a class or a probability distribution about the class, indicating that the decision tree has classified the dataset into a particular class or a particular probability distribution (which may be biased towards a subset of classes). For example, each class is represented by a label (e.g., an extended time label), such as... Figure 2A The numbers 1, 2, and 3 in the text.

[0040] like Figure 1As can be seen, the seven nodes (nodes A and B) of decision tree 60 represent one or more attributes (e.g., features) A ​​and B. A and B are attributes associated with the user (e.g., distinct attributes). The node representing attribute A can be called node A. The node representing attribute B can be called node B. The child nodes of node A are connected to node A through branches A1 to A7 representing the corresponding labels of attribute A. In other words, in the acquired historical data, attribute A is given any one of A1 to A7 as a value or label. Node B can be considered a child node of node A (the root node of decision tree 60).

[0041] Decision tree 60 includes branches representing A1 to A7 and B1 to B2. For example, labels A1 to A7 are associated with attribute A. Labels A1 to A7 can be considered unique labels (e.g., each label may be different). In other words, the branches representing attributes A1 to A7 can be considered as branches originating from node A. The branches representing attributes A1 to A7 can be considered as the first branches of decision tree 60.

[0042] Labels B1 to B2 can be viewed as branches of node B. In other words, the branches representing attributes B1 to B2 can be considered as branches originating from node B. The branch representing attribute B1 can be considered the first main branch of decision tree 60. The branch representing attribute B2 can be considered the first branch of decision tree 60.

[0043] Figure 2B This illustrates a representation of an example decision tree 80 according to this disclosure. The structure of decision tree 80 is similar to... Figure 2A The structure of decision tree 60 shown is similar. Decision tree 80 is associated with a decision tree classifier. In other words, decision tree 80 includes internal nodes, branches, and leaf nodes. Decision tree 80 is generated, for example, based on historical data associated with users (shown in Table 1 below).

[0044] Table 1 Table 1 shows historical data associated with users. For example, each row in Table 1 is associated with a historical shipment associated with the user. Table 1 includes attributes and extension time parameters. The attributes shown in Table 1 are contract codes (e.g., 47 and 57) and countries (e.g., Denmark, India, etc.). As shown in Table 1, each contract code can be associated with one or more shipments. The extension time parameters shown in Table 1 are values ​​between 1 and 3. For example, the extension time parameters shown in Table 1 can be viewed as time slots, such as extension time parameters.

[0045] The root node of decision tree 80 (e.g., Figure 2ANode A) represents the country of shipment associated with the user (e.g., the destination country). For example, the country is an attribute of the shipment associated with the user. Figure 2A The node A shown is related to the representation. Figure 2B The root node corresponds to the national attribute.

[0046] Decision tree 80 includes connections to the root node representing the national attribute (e.g., Figure 2A Node A) and internal nodes representing contract code attributes (e.g., Figure 2B The branch of node B) (e.g., Figure 2A Branches A1 to A7). A branch connecting the root node and internal nodes can be considered the first branch. For example... Figure 2B As shown, the first branch can represent a country corresponding to the historical data provided in Table 1 (e.g., India, the United Kingdom, Denmark, the United States, Japan, Indonesia, China, etc.). It should be noted that for the country code UK, the decision tree classifier directly provides option 2 as the perfect classification option. For example, the United Kingdom shown in Table 1 is associated with only one extended time parameter (slot 2). This is shown in decision tree 80 as a branch directly connecting the root node representing the country attribute to the leaf node representing the extended time period (slot 2) (e.g., the first branch A7).

[0047] Decision tree 80 includes internal nodes representing the contract code of the shipment associated with the user (e.g., Figure 2A (Internal node B). The contract code can be viewed as the contract code (e.g., contract number) of the contract (e.g., agreement) performed on which the shipment (e.g., historical shipment) associated with the user is based.

[0048] Decision tree 80 includes connections between internal nodes representing contract codes (e.g., Figure 2A The internal node (node ​​B) and the leaf node (branch B1 to B2) representing the extended time parameter (e.g., time slot 1, 2, or 3) are connected. The branch connecting the internal node and the leaf node can be considered a secondary branch. In some examples, each internal node may be associated with one or more sub-branches that connect the internal node to its child nodes (such as leaf nodes, e.g., nodes representing the extended time parameter). See Table 1 and... Figure 2B As shown, two labels (47 and 57) are associated with the contract code. In other words, the value of the contract code can be either 47 or 57. The internal nodes of decision tree 80 include two sub-branches, each connecting the internal node to its child nodes. The second branch associated with each internal node of decision tree 80 represents a given contract code from historical data, for example, 47 or 57.

[0049] The decision tree 80 includes one or more leaf nodes representing one or more corresponding time extension parameters. For example, the decision tree 80 includes leaf nodes representing the following time extension parameters: slot 1, 2, or 3. The time extension parameter of the leaf node can be, for example, considered as the time extension parameter of the classification.

[0050] For example, predicting extended time periods based on historical data (e.g., Table 1) and booking requests includes applying a predictive model (e.g., a decision tree classifier, such as decision tree 80) to the historical data.

[0051] In some examples, predicting extended time periods based on historical data (e.g., Table 1) and booking requests includes determining whether the historical data provides the same extended time parameter for all data elements of the historical data. In other words, for example, the device determines whether all rows of the historical data (e.g., in Table 1) have the same extended time parameter. For example, after determining that all rows of the historical data (e.g., rows of the table (e.g., Table 1)) have the same extended time parameter, the device sets the current node as a leaf node. In other words, for example, when determining that all rows of the historical data (e.g., in Table 1) have the same extended time parameter, all rows belong to the same category. In other words, when determining that the extended time parameter of all data elements of the historical data is the same, the historical data can be considered to have been classified, e.g., perfectly classified. For example, when determining that the extended time parameter of all data elements of the historical data is the same (e.g., when the historical data has been classified), the extended time parameter associated with the attribute is selected as the predicted extended time period. In some examples, the method includes generating a leaf node representing the extended time parameter after determining that the extended time parameter of all data elements of the historical data is the same.

[0052] In some examples, the method includes checking whether the historical data (e.g., Table 1) provides at least one attribute after determining that all data elements of the historical data (e.g., rows of a table, such as rows of Table 1) do not share the same extended time parameter. For example, when all data elements (country and contract code) shown in Table 1 are not associated with the same extended time parameter, the electronic device determines whether the historical data provides at least one attribute. For example, the electronic device verifies the count of parameters in the historical data (such as, for example, columns of a table (e.g., Table 1) retrieved from a database).

[0053] In some examples, when it is determined that the historical data does not include attributes, the electronic device sets the current node (e.g., the root node or an internal node) as a leaf node. For instance, a leaf node represents the most frequently occurring extension time parameter in the historical data. In other words, the current node is given the most frequently occurring category as its label. For example, when the historical data includes extension time parameters but does not include attributes (e.g., country and / or contract code), the most frequently occurring extension time parameter can be selected as the predicted extension time parameter.

[0054] In some examples, when at least one attribute is provided in the historical data, the method includes calculating the information gain of that at least one attribute. In some examples, the historical data includes one or more attributes, and the information gain of each attribute is calculated.

[0055] In some examples, the method includes selecting a first attribute from one or more attributes that provides the highest information gain. In other words, for example, historical data is partitioned into subsets based on the attribute with the highest information gain (e.g., the attribute with the highest information gain among the attributes of historical data).

[0056] Information gain can be viewed as an indicator of the effectiveness of a given attribute (e.g., a feature) in classifying a target category. The attribute with the highest information gain can be considered the best-performing attribute (e.g., for classifying historical data).

[0057] For example, the information gain of an attribute of a given dataset can be expressed as: Where p(I) represents the probability of the total number of distinct values ​​occurring relative to the total number of values ​​occurring, S represents the set (e.g., the entire dataset of historical data), A represents the attribute, and n represents the number of unique output values. The entropy of a dataset is an indicator of the disorder and / or impurity in the target attribute (e.g., feature) of the dataset.

[0058] Where I represents a random variable that indicates the occurrence of a unique label for the extended time period (e.g., a time slot) of the output of the target attribute.

[0059] In some examples, the method includes generating a decision tree based on the attribute with the highest information gain (e.g., decision tree 80). In other words, the method may include generating a decision tree where the root node represents the attribute with the highest calculated information gain among the information gains calculated for the attributes of historical data (referred to as the first attribute). In some examples, the method includes generating a node representing the attribute with the highest information gain (referred to as the first attribute) (e.g., a child node of an internal node, such as a child node of the root node).

[0060] In some examples, the method includes generating one or more nodes of a decision tree. For example, generating one or more (e.g., each) nodes of a decision tree may include determining which attribute has the highest information gain. In one or more examples, predicting an extended time period includes, after determining that the historical data includes at least one attribute, selecting the first attribute among one or more attributes that provides the highest information gain as the root node for the decision. For example, the first attribute of the root node of decision tree 80 is country. For decision tree 80, country can be considered the attribute with the highest information gain (e.g., the first attribute). For example, the attribute of a child node of the root node is contract code. Contract code (in decision tree 80) has the second highest information gain. In other words, the attribute with the second highest information gain can be considered the second attribute (e.g., a child node of the root node). For example, representing... Figure 2B The internal nodes of the contract code shown (e.g., first internal non-root nodes, such as children of the root node) represent second attributes (e.g., attributes with the second highest information gain, such as contract codes as shown in Table 1). Children of the root node can be considered as first internal non-root nodes. Children of the first internal non-root node can be considered as second internal non-root nodes.

[0061] In other words, the root node's attribute can be, for example (historical data), the attribute with the highest calculated information gain. The attribute of the first internal non-root node can be, for example (historical data), the attribute with the second highest calculated information gain among one or more attributes.

[0062] In some examples, when the historical data includes a third attribute whose calculated information gain is lower than that of the second attribute, the electronic device generates a child node representing the internal non-root node of the third attribute.

[0063] This process can be repeated until all attributes of the historical data are exhausted and all data elements of the historical data are classified, for example, belonging to the same category.

[0064] Figure 3A -B illustrates a flowchart of an exemplary method 100 for predicting an extended time period, performed by an electronic device according to this disclosure. Method 100 is performed by an electronic device, such as those disclosed herein, such as... Figure 5 Electronic device 300.

[0065] Method 100 includes obtaining, in S102, a shipment booking request associated with the user. For example, method 100 includes receiving and / or retrieving a shipment booking request associated with the user from a booking platform and / or the user's device. For example, the booking request may be considered a user-specific request for a shipment associated with the user. For example, the booking request may be considered a request for an extended time period.

[0066] Method 100 includes retrieving historical data associated with the user in S106 based on a booking request. In some examples, retrieving historical data associated with the user in S106 based on a booking request includes, for example, retrieving and / or receiving historical data from local storage devices and / or remote storage devices. In some examples, retrieving historical data associated with the user in S106 based on a booking request includes retrieving data from remote storage devices (e.g., from a database, such as...) Figure 1 Database 16) Obtaining historical data. In one or more example methods, obtaining historical data in S106 includes obtaining one or more of the following in S106A: one or more attributes, one or more tags associated with each attribute, a count of each attribute, and one or more extension time parameters indicating one or more previous extension time periods associated with the corresponding one or more attributes. In some examples, obtaining one or more of the following in S106A: one or more attributes, one or more tags associated with each attribute, a count of each attribute, and one or more extension time parameters indicating one or more previous extension time periods associated with the corresponding one or more attributes includes determining (e.g., generating) one or more of the following: one or more attributes, one or more tags associated with each attribute, a count of each attribute, and one or more extension time parameters indicating one or more previous extension time periods associated with the corresponding one or more attributes. It is conceivable that, at the initial stage, the initial historical data associated with the user includes booking data (e.g., historical booking data). For example, in this case, the historical data is generated by the electronic device by indexing the initial historical data to determine one or more attributes, one or more tags associated with each attribute, a count of each attribute, and / or one or more extension time parameters. For example, historical data is generated by indexing initial historical data associated with users.

[0067] Method 100 includes predicting the extended time period S108 based on historical data and booking requests by applying a predictive model to historical data. In some examples, the extended time period is predicted based on one or more of the following: one or more attributes, one or more labels associated with each attribute, a count of each attribute, one or more extension time parameters indicating one or more previous extended time periods associated with the corresponding one or more attributes, and booking requests. In other words, in some examples, the predictive model is applied to one or more of the following: one or more attributes, one or more labels associated with each attribute, a count of each attribute, one or more extension time parameters indicating one or more previous extended time periods associated with the corresponding one or more attributes, and booking requests.

[0068] Method 100 includes providing a booking response (S110) to the user based on the predicted extended time period. In some examples, providing a booking response (S110) to the user based on the predicted extended time period may include providing the user with the predicted extended time period. For example, the booking response may include the predicted extended time period. The booking response may be considered as a response to a booking request for a shipment associated with the user. The booking response may include, for example, the predicted extended time parameter.

[0069] In one or more example methods, the predictive model is a machine learning predictive model. In one or more example methods, the machine learning predictive model is a decision tree classifier. A decision tree classifier can be viewed as a machine learning classifier used to classify information (e.g., historical data associated with a user, such as initial historical data associated with the user). A decision tree classifier can have a tree structure, where internal nodes can be represented by attributes (e.g., features). For example, each internal node represents a decision based on a given attribute. For example, the leaf nodes of the decision classifier's tree structure represent the results (e.g., outputs) of the decision tree classifier. In other words, internal nodes correspond to decision rules based on given attributes, while leaf nodes represent the resulting classification. A decision tree classifier can be configured to generate a decision tree based on the highest information gain. For example, during classification, a user input instance (e.g., product code, user code, contract code, and / or country) can be guided down the tree according to decision rules until it reaches a leaf node (e.g., an extension time parameter). Figure 2A -B illustrates a decision tree classifier according to this disclosure. Attributes can be considered as decision variables and / or decision parameters of the decision tree classifier.

[0070] In one or more example methods, predicting the S108 time period includes training an S108A decision tree classifier based on historical data. In one or more example methods, training an S108A decision tree classifier based on historical data includes classifying the historical data S108AA using a decision tree classifier (e.g., any of those in S108B-I).

[0071] In one or more example methods, predicting the extended time period in S108 includes determining whether the historical data in S108B provides the same extended time parameter for all data elements of the historical data. In other words, the historical data includes one or more data elements (e.g., Table 1). In one or more example methods, predicting the extended time period in S108 includes selecting the value of the same extended time parameter in S108C as the predicted extended time period after determining that the historical data provides the same extended time parameter for all data elements of the historical data. For example, when the historical data provides the same extended time parameter (e.g., value and / or time period) for all data elements, the extended time period provided in the extended time parameter of the historical data is the extended time period selected as the predicted extended time period and provided to the scheduled response.

[0072] In one or more example methods, predicting the S108 extended time period includes determining, after determining that the historical data does not provide the same extended time parameter for all data elements of the historical data, whether the historical data includes at least one attribute (S108D). In some examples, the historical data includes multiple attributes, such as a first attribute, a second attribute, and optionally a third attribute, and optionally a fourth attribute, etc.

[0073] In one or more example methods, predicting the S108 extension time period involves selecting the most frequently occurring extension time parameter in S108E as the predicted extension time period after determining that the historical data does not include at least one attribute. For example, the most frequently occurring extension time parameter in the historical data is the parameter that appears most frequently in the historical data. Figure 2B As shown in the image.

[0074] In one or more example methods, predicting the extended time period of S108 includes determining the information gain of each attribute (e.g., one or more attributes) in S108F after determining that the historical data includes at least one attribute (e.g., one or more attributes). Figure 2B As shown in the image.

[0075] In one or more example methods, predicting the extended time period of S108 involves selecting a first attribute from one or more attributes of S108G that provides the highest information gain after determining that the historical data includes at least one attribute. The information gain is shown, for example, by Equation (1).

[0076] In one or more example methods, predicting the extended time period S108 includes, after determining that historical data includes at least one attribute, generating a first decision tree S108H based on the first attribute, the first decision tree having the first attribute as the root node and one or more leaf nodes representing one or more of the predicted extended time periods. An example first decision tree can be considered as... Figure 2A-B shows a decision tree with a root node, non-root internal nodes, and leaf nodes. The first decision tree can be viewed as a decision tree with the first attribute as its root node. In other words, the first decision tree can be viewed as a decision tree where the root node represents the attribute with the highest information gain. For example, the second decision tree can be viewed as a decision tree with subtrees like the first decision tree, thus having the same root node as the first decision tree.

[0077] In one or more example methods, predicting the extended time period S108 involves classifying the historical data using a first decision tree S108I after determining that the historical data includes at least one attribute. This can result in additional internal non-root nodes appearing in the first decision tree, as for... Figure 2A -B is shown.

[0078] In one or more example methods, predicting the extended time period S108 involves repeating one or more of steps S108B to S108I until all attributes of one or more attributes of the historical data have been processed. In some examples, classifying the historical data using a decision tree classifier S108AA involves repeating one or more of steps S108B to S108I, for example, until all attributes of one or more attributes of the historical data have been processed.

[0079] In one or more example methods, method 100 includes determining, in step S104, whether historical data can be retrieved from the storage device. In one or more example methods, method 100 includes retrieving historical data associated with a user from the storage device in step S106B after determining that historical data can be retrieved from the storage device. In one or more example methods, method 100 includes prohibiting the retrieval of historical data (e.g., exiting method 100) after determining that historical data associated with a user cannot be retrieved from the storage device.

[0080] In one or more example methods, providing a booking response to the user in S110 based on the predicted extended time period includes causing S110A to display a user interface object representing the predicted extended time period. In some examples, the user device may include a display configured to display the user interface object representing the predicted extended time period in response to receiving the booking response provided in S110.

[0081] Figure 4 This is an example representation based on the example output of this disclosure. For example, Figure 4 A user interface is shown that can be displayed on a user device, including a display, in response to receiving a subscription response disclosed herein.

[0082] Figure 4A user interface object 92 is shown, which represents a table 90 of example predicted extension time periods and example time extension values ​​provided in the booking response. For example, providing a booking response can be considered as causing electronic devices (e.g., Figure 5 The interface 303 of the electronic device 300 shown displays a user interface object 92 representing the predicted extended time period. In other words, the user interface object 92 represents the predicted extended time period. For example, the user interface object 92 is a drop-down menu that provides multiple extended time periods predicted according to this disclosure. For example, the user interface object 92 can enable a user to select an extended time period, for example, based on the predicted extended time period. In some examples, the predicted extended time period can be regarded as a recommended predicted extended time period.

[0083] User interface object 92 indicates that the predicted extension period is +4 days (e.g., an extension of 4 days from the existing period). In other words, the electronic device can be configured to add 4 days to a free period of 1 to 12 days (e.g., 12 days free) (e.g., based on user input).

[0084] exist Figure 4 In the example, user interface object 92 indicates the value (e.g., price) generated when the user selects an extended time period (e.g., paid by the port authority and / or the shipper). For example, the generated value is displayed in a given currency (e.g., as shown in the example). Figure 4 The Indian Rupee (INR) is shown.

[0085] Table 90, Extended Time Value, includes information indicating the value (e.g., price) of extended time periods. For example, time periods 1 to 12 can be considered D&D time periods without incurring additional costs. Time periods 13 to 15 and 15+ can be considered extended time periods. The values ​​in the table (0, 120.00, and 350.00) are in UAE Dirham (AED) currency (any currency may be used).

[0086] Figure 5 A block diagram of an exemplary electronic system 300 according to the present disclosure is shown. The electronic device 300 includes a memory circuitry 301, a processor circuitry 302, and an interface 303. The electronic device 300 is configured to perform... Figure 3A Any of the methods disclosed in -B. In other words, the electronic device 300 is configured to predict the extended time period.

[0087] In some examples, electronic device 300 is an extended time prediction electronic device. In some examples, electronic device 300 is an extended time prediction electronic system. In some examples, electronic device 300 is an extended time predictor.

[0088] Electronic device 300 is configured (e.g., via memory circuitry 301 and / or interface 303) to acquire shipping reservation requests associated with a user.

[0089] Electronic device 300 is configured (e.g., via memory circuitry 301 and / or interface 303) to retrieve historical data associated with a user based on a subscription request.

[0090] Electronic device 300 is configured to predict extended time periods based on historical data and booking requests by applying a predictive model to historical data (e.g., via processor circuitry 302) through memory circuitry 301 and / or processor circuitry 302.

[0091] Electronic device 300 is configured to provide a pre-booked response to a user based on a predicted extended time period (e.g., via processor circuitry 302 and / or interface 303).

[0092] Processor circuit 302 is optionally configured to execute Figure 3A Any of the operations disclosed in -B (such as one or more of the following: S102, S104, S106, S106A, S106B, S108, S108A, S108B, S108C, S108CA, S108D, S108E, S108F, S108G, S108H, S108I, S110, S110A). The operation of electronic device 300 may be embodied in the form of executable logic routines (e.g., lines of code, software programs, etc.) stored on a non-transitory computer-readable medium (e.g., memory circuitry 301) and executed by processor circuitry 302.

[0093] Furthermore, the operation of electronic device 300 can be considered as a method configured to be performed by electronic device 300. Additionally, while the described functions and operations can be implemented in software, such functions can also be implemented via dedicated hardware or firmware, or some combination of hardware, firmware, and / or software.

[0094] The memory circuit system 301 may be one or more of a buffer, flash memory, hard disk drive, removable media, volatile memory, non-volatile memory, random access memory (RAM), or other suitable devices. In a typical arrangement, the memory circuit system 301 may include non-volatile memory for long-term data storage and volatile memory used as system memory for the processor circuit system 302. The memory circuit 301 exchanges data with the processor circuit 302 via a data bus. Control lines and an address bus may also exist between the memory circuit 301 and the processor circuit 302. Figure 5(Not shown in the image). The memory circuitry 301 is considered a non-transitory computer-readable medium.

[0095] The memory circuit system 301 can be configured to store reservation requests, historical data, extended time periods, predictive models, reservation responses, one or more attributes, one or more labels associated with each attribute, counts of each attribute, one or more extended time parameters, information gain, and / or user interface objects in a portion of the memory.

[0096] The implementation of the methods and products (electronic devices) according to this disclosure is set forth in the following terms: Clause 1. A method performed by an electronic device, the method comprising: Retrieve shipping reservation requests associated with the user; Based on the booking request, obtain the historical data associated with the user; Based on this historical data and the booking request, a predictive model is applied to the historical data to predict an extended time period; and Based on the predicted extended period, a booking response is provided to the user.

[0097] Clause 2. The method described in Clause 1, wherein the prediction model is a machine learning prediction model.

[0098] Clause 3. The method according to any one of the preceding clauses, wherein the machine learning prediction model is a decision tree classifier.

[0099] Clause 4. The method according to any one of the preceding clauses, wherein obtaining the historical data includes obtaining one or more of the following: one or more attributes, one or more tags associated with each attribute, a count of each tag associated with the attribute, and one or more extension time parameters indicating one or more previous extension time periods associated with the corresponding one or more attributes.

[0100] Clause 5. The method according to any one of Clauses 3 to 4, wherein predicting the extended time period includes training the decision tree classifier based on the historical data.

[0101] Clause 6. The method according to Clause 5, wherein training the decision tree classifier based on the historical data includes classifying the historical data using the decision tree classifier.

[0102] Clause 7. The method according to any one of the preceding clauses, wherein predicting the extended time period includes: - Determine whether the historical data provides the same extended time parameter for all data elements of the historical data; and - Once it is determined that the historical data provides the same extended time parameter for all data elements of the historical data, the value of the same extended time parameter is selected as the extended time period for the prediction.

[0103] Clause 8. The method according to any one of the preceding clauses, wherein predicting the extended time period includes: - Once it is determined that the historical data does not provide the same extended time parameter for all data elements of the historical data, determine whether the historical data includes at least one attribute.

[0104] Clause 9. The method according to any one of the preceding clauses, wherein predicting the extended time period includes: - Once it is determined that the historical data does not include at least one attribute, the most frequently occurring extension time parameter is selected as the extension time period for the prediction.

[0105] Clause 10. The method according to any one of the preceding clauses, wherein predicting the extended time period includes determining that the historical data includes at least one attribute: Determine the information gain for each attribute; Select the first attribute from the one or more attributes that provides the highest information gain; and A first decision tree is generated based on the first attribute, which has a first attribute as the root node and one or more leaf nodes representing one or more predicted extended time periods. The first decision tree is used to classify the historical data.

[0106] Clause 11. The method according to any one of the preceding clauses, wherein predicting the extended time period includes repeating one or more of the steps until all attributes of one or more attributes of the historical data have been processed.

[0107] Clause 12. The method according to any one of the preceding clauses, the method comprising determining whether the historical data can be retrieved from the storage device.

[0108] Clause 13. The method according to any one of the preceding clauses, wherein providing the booking response to the user based on the predicted extended time period includes causing a user interface object to be displayed representing the predicted extended time period.

[0109] Clause 14. An electronic device comprising memory circuitry, processor circuitry, and an interface, wherein the electronic device is configured to perform any of the methods according to any one of Clauses 1 to 13.

[0110] Clause 15. A computer-readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by an electronic device, cause the electronic device to perform any of the methods described in Clauses 1 to 13.

[0111] The use of terms such as "first," "second," "third," and "fourth," "primary," "secondary," and "auxiliary," etc., does not imply any specific order, but is included to identify individual elements. Furthermore, the use of terms such as "first," "second," "third," and "fourth," "primary," "secondary," and "auxiliary," etc., does not indicate any order or importance, but is used to distinguish one element from another. Note that the use of the terms "first," "second," "third," and "fourth," "primary," "secondary," and "auxiliary," etc., here and elsewhere, is solely for labelling purposes and is not intended to indicate any specific spatial or temporal order. Moreover, the labeling of a first element does not imply the existence of a second element, and vice versa.

[0112] It should be understood that Figures 1 to 5 This includes some circuits or operations shown with solid lines and some circuits or operations shown with dashed lines. The circuits or operations included in the solid lines are those included in the most broad exemplary embodiments. The circuit systems or operations included in the dashed lines are exemplary embodiments that can be included in or part of the circuit systems or operations of the solid-line example embodiments, or are further circuit systems or operations that can be taken in addition to the circuit systems or operations of the solid-line example embodiments. It should be understood that these operations do not need to be performed in the order presented. Furthermore, it should be understood that not all operations need to be performed. The exemplary operations can be performed in any order and in any combination.

[0113] It should be noted that the word "including" does not necessarily exclude the existence of other elements or steps besides those listed.

[0114] It should be noted that the words "one" or "a kind" preceding an element do not preclude the existence of multiple such elements.

[0115] It should also be noted that any reference numerals in the drawings do not limit the scope of the claims, exemplary embodiments may be implemented at least in part by both hardware and software, and several “components,” “units,” or “devices” may be represented by the same hardware article.

[0116] The various exemplary methods, devices, nodes, and systems described herein are described in the general context of method steps or processes. In one aspect, these method steps or processes may be implemented by a computer program product embodied in a computer-readable medium, including computer-executable instructions, such as program code, that are executed by a computer in a networked environment. Computer-readable media may include removable and non-removable storage devices, including but not limited to read-only memory (ROM), random access memory (RAM), optical disc (CD), digital versatile disc (DVD), etc. Generally, program circuitry may include routines, programs, objects, components, data structures, etc., that perform a specified task or implement a particular abstract data type. Computer-executable instructions, associated data structures, and program circuitry represent examples of program code for performing steps of the methods disclosed herein. Specific sequences of such executable instructions or associated data structures represent examples of corresponding actions for implementing the functionality described in such steps or processes.

[0117] Although features have been shown and described, it should be understood that they are not intended to limit the scope of the claimed disclosure, and it will be apparent to those skilled in the art that various changes and modifications can be made without departing from the scope of the claimed disclosure. Therefore, this specification and accompanying drawings should be considered illustrative rather than restrictive. The claimed disclosure is intended to cover all alternatives, modifications, and equivalents.

Claims

1. A method performed by an electronic device, the method comprising: Retrieve shipping reservation requests associated with the user; Based on the booking request, obtain historical data associated with the user; Based on the historical data and the booking request, an extended time period is predicted by applying a predictive model to the historical data. as well as Based on the predicted extended time period, a booking response is provided to the user.

2. The method according to claim 1, wherein the prediction model is a machine learning prediction model.

3. The method according to any one of the preceding claims, wherein the machine learning prediction model is a decision tree classifier.

4. The method according to any one of the preceding claims, wherein obtaining the historical data includes obtaining one or more of the following: one or more attributes, one or more tags associated with each attribute, a count of each tag associated with the attribute, and one or more extension time parameters, the one or more extension time parameters indicating one or more previous extension time periods associated with the corresponding one or more attributes.

5. The method according to any one of claims 3 to 4, wherein predicting the extended time period includes training the decision tree classifier based on the historical data.

6. The method of claim 5, wherein training the decision tree classifier based on the historical data comprises classifying the historical data using the decision tree classifier.

7. The method according to any one of the preceding claims, wherein predicting the extended time period comprises: - Determine whether the historical data provides the same extended time parameter for all data elements of the historical data; as well as - Once it is determined that the historical data provides the same extended time parameter for all data elements of the historical data, the value of the same extended time parameter is selected as the predicted extended time period.

8. The method according to any one of the preceding claims, wherein predicting the extended time period comprises: - Once it is determined that the historical data does not provide the same extended time parameter for all data elements of the historical data, determine whether the historical data includes at least one attribute.

9. The method according to any one of the preceding claims, wherein predicting the extended time period comprises: - Once it is determined that the historical data does not include at least one attribute, the most frequently occurring extension time parameter is selected as the predicted extension time period.

10. The method according to any one of the preceding claims, wherein predicting the extended time period includes determining that the historical data includes at least one attribute: Determine the information gain for each attribute; Select a first attribute from the one or more attributes that provides the highest information gain; and A first decision tree is generated based on the first attribute, the first decision tree having the first attribute as the root node and one or more leaf nodes representing one or more predicted extended time periods. The historical data is classified using the first decision tree.

Images