Method, device, computer equipment and storage medium for determining item description words

Abstract

The present disclosure proposes a method, device, computer equipment and storage medium for determining item description words, and relates to the field of computer technology. The method includes: determining a target system sentence, an associated input sentence, and an item corresponding to the target system sentence contained in a conversation log; determining the candidate words contained in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in a preset descriptive word library; determining the corresponding relationship between each item and the associated input sentence, and the candidate words contained in the associated input sentence; determining the descriptive words contained in each candidate word corresponding to each item according to the occurrence probability of the candidate word corresponding to each item. Thus, by automatically extracting and processing the conversation log, the corresponding descriptive words can be generated, the generation efficiency of the descriptive words is higher, and the obtained descriptive words are more accurate and reliable.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method, device, computer equipment and storage medium for determining an item description word. Background Art

[0002] With the rapid development of Internet technology, various e-commerce platforms provide a variety of online commodity trading channels, which greatly facilitates people's work and life. In the e-commerce consultation dialogue system, the items that the user wants to know can be determined by searching for the corresponding descriptive words of each item according to the sentence input by the user.

[0003] In the related art, it is usually necessary to manually determine the descriptive words corresponding to each item. Due to the large number of goods, a lot of labor costs are required. At the same time, the efficiency of manual summarization is not high.

[0004] Therefore, how to quickly and accurately determine the description words of the items is crucial. Summary of the invention

[0005] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.

[0006] The first embodiment of the present disclosure provides a method for determining an item description word, including:

[0007] Determine the target system statement, the associated input statement, and the item corresponding to the target system statement contained in the conversation log;

[0008] Determining candidate words included in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in a preset descriptive word library;

[0009] Determining the correspondence between each item and the candidate word according to the correspondence between each item and the associated input sentence and the candidate words included in the associated input sentence;

[0010] According to the occurrence probability of the candidate words corresponding to each of the items, the descriptive words included in each of the candidate words corresponding to each of the items are determined.

[0011] The second aspect of the present disclosure provides a device for determining an item description word, including:

[0012] A first determination module, used to determine the target system statement, the associated input statement, and the item corresponding to the target system statement contained in the conversation log;

[0013] A second determination module, configured to determine the candidate words included in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in a preset descriptive word library;

[0014] A third determination module, configured to determine a correspondence between each of the items and the candidate words according to the correspondence between each item and the associated input sentence and the candidate words included in the associated input sentence;

[0015] The fourth determination module is used to determine the descriptive words included in each candidate word corresponding to each of the items according to the occurrence probability of the candidate word corresponding to each of the items.

[0016] The third aspect of the present disclosure provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for determining the item description word provided in the first aspect of the present disclosure is implemented.

[0017] The fourth aspect of the present disclosure provides a non-temporary computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for determining an item descriptor as provided in the first aspect of the present disclosure.

[0018] The fifth aspect of the present disclosure provides a computer program product. When an instruction processor in the computer program product executes, the method for determining the item description word provided in the first aspect of the present disclosure is performed.

[0019] The method, device, computer equipment and storage medium for determining the item description words provided by the present disclosure can first determine the target system sentence, the associated input sentence and the item corresponding to the target system sentence contained in the dialogue log, then determine the candidate words contained in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in the preset descriptive word library, then determine the corresponding relationship between each item and the associated input sentence and the candidate words contained in the associated input sentence, and then determine the descriptive words contained in each candidate word corresponding to each item according to the occurrence probability of the candidate word corresponding to each item. Thus, by automatically extracting and processing the dialogue log, the corresponding descriptive words can be generated, the generation efficiency of the descriptive words is higher, and the obtained descriptive words are more accurate and reliable.

[0020] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0022] Figure 1A schematic diagram of a flow chart of a method for determining an item description word provided by an embodiment of the present disclosure;

[0023] Figure 2 A flowchart of a method for determining an item description word provided by another embodiment of the present disclosure;

[0024] Figure 3 A flowchart of a method for determining an item description word provided by another embodiment of the present disclosure;

[0025] Figure 4 A flowchart of a method for determining an item description word provided by another embodiment of the present disclosure;

[0026] Figure 5 A schematic diagram of the structure of a device for determining an item description word provided by an embodiment of the present disclosure;

[0027] Figure 6 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0028] Embodiments of the present disclosure are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0029] The following describes a method, apparatus, computer device, and storage medium for determining an item description word according to an embodiment of the present disclosure with reference to the accompanying drawings.

[0030] Figure 1 A schematic diagram of a flow chart of a method for determining an item description word provided in an embodiment of the present disclosure.

[0031] The embodiment of the present disclosure takes the method for determining an item descriptor as an example in which the method is configured in a device for determining an item descriptor. The device for determining an item descriptor can be applied to any computer device so that the computer device can perform the function of determining an item descriptor.

[0032] Among them, the computer device can be a personal computer (PC), a cloud device, a mobile device, etc. The mobile device can be, for example, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, a car device, etc., which are hardware devices with various operating systems, touch screens and / or display screens.

[0033] For the convenience of explanation, the determining device of the item description word can be simply referred to as "determining device".

[0034] like Figure 1 As shown, the method for determining the item description word may include the following steps:

[0035] Step 101, determining the target system statement, the associated input statement, and the object corresponding to the target system statement contained in the dialogue log.

[0036] The conversation log may include multiple statements, such as system statements, input statements, etc., which is not limited in the present disclosure.

[0037] For example, the conversation log may contain sentences such as: recommend a mobile phone for the elderly; what is the approximate price; about 2,000 yuan; do you have a favorite brand; no. OK, please wait; please click the link below to view the recommended item details: XX mobile phone ￥1,899... etc., and this disclosure does not limit this.

[0038] Optionally, a system statement in the conversation log that is used to point to the page where the item is located can be determined as the target system statement.

[0039] For example, in the above example, "Please click the link below to view the recommended item details: XX mobile phone ￥1899..." is a system statement that can point to the page where the item is located, so it can be used as the target system statement.

[0040] Optionally, the item corresponding to the target system statement may be determined based on the description information of the page where the item pointed to by the target system statement is located.

[0041] For example, the description information of the page where the item is located is: XX brand Y series 9th generation mobile phone, 100 million pixels, fast charging support, ￥1899, etc. Then, based on the description information, it can be determined that the item corresponding to the target system statement is: XX mobile phone Y series 9th generation mobile phone.

[0042] Optionally, a preset number of input statements received before the target system statement may be determined as associated input statements.

[0043] The preset number may be a value set in advance, such as 3, 5, etc., and the present disclosure does not limit this.

[0044] For example, the preset number is 3. Then the input sentences in the above example, "recommend a mobile phone for the elderly", "about 2000", "no", can be determined as related input sentences.

[0045] It should be noted that the above examples are merely illustrative and cannot be used as limitations on the statements, target system statements, associated input statements, and items corresponding to the target system statements contained in the conversation log in the embodiments of the present disclosure.

[0046] It is understandable that at the same time, there may be multiple target system statements in the conversation log, and each target system statement, the object corresponding to each target system statement, and the associated input statement, etc. can be determined separately, and the present disclosure does not limit this.

[0047] It is understandable that the content in the conversation log is more comprehensive and closer to the user's expression and feelings. Therefore, in the embodiment of the present disclosure, by mining and processing the conversation log, the determined content can be more accurate and more in line with user needs.

[0048] Step 102: Determine the candidate words included in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in a preset descriptive word library.

[0049] The candidate words included in the associated input sentence may be one or more than one, and the present disclosure does not limit this.

[0050] It is understandable that the contents of detail pages of a large number of items may be parsed in advance, and the processed results may be used as descriptive words in a preset descriptive word library.

[0051] In addition, the associated input sentence can be parsed first to obtain the participles it contains, and the participles irrelevant to the item description can be eliminated, and then the remaining participles are compared with the description words in the preset description word library to determine the candidate words contained in the associated input sentence.

[0052] For example, by parsing the associated input sentence, the determined segment words are: hello, fan, cabinet type, high power, and movable. Among them, "hello" is irrelevant to the description of the item, so it can be removed first, and then the remaining segment words "fan, cabinet type, high power, and movable" are matched with each descriptive word in the preset description word library. If the descriptive words in the preset description word library do not include "removable", that is, the matching degree is zero, it can be determined that "removable" is a candidate word included in the associated input sentence.

[0053] It should be noted that the above examples are merely illustrative and cannot be used as limitations on the associated input sentences and the candidate words contained therein in the embodiments of the present disclosure.

[0054] In the disclosed embodiment, the associated input sentence may be parsed first and then matched with each descriptive word in a preset descriptive word library to determine the corresponding matching degree, and then the corresponding candidate word may be determined based on each matching degree, thereby improving the accuracy and reliability of the determined candidate word.

[0055] Step 103 , determining the correspondence between each item and the candidate word according to the correspondence between each item and the associated input sentence and the candidate words included in the associated input sentence.

[0056] It is understandable that there may be one or more associated input sentences in a conversation log; different associated input sentences may correspond to the same item or different items, etc.; the candidate words contained in each associated input sentence may be the same or different, etc., and the present disclosure does not limit this.

[0057] For example, associated input sentence 1 corresponds to item 1, associated input sentence 2 corresponds to item 2, and associated input sentence 3 corresponds to item 1, and the candidate word of associated input sentence 1 is A, the candidate word of associated input sentence 2 is B, and the candidate word of associated input sentence 3 is C. Then it can be determined that item 1 corresponds to candidate word A and candidate word C, and item 2 corresponds to candidate word B.

[0058] It should be noted that the above examples are merely illustrative and cannot be used as a limitation on the correspondence between objects, related input sentences, and candidate words in the embodiments of the present disclosure.

[0059] Step 104 , determining the descriptive words included in each candidate word corresponding to each item according to the occurrence probability of the candidate word corresponding to each item.

[0060] The descriptive words may be any words that can be used to describe an item, such as: for the elderly, to wear in winter, can slide, feels good, etc., which are not limited in the present disclosure.

[0061] Optionally, the candidate word with the highest probability of occurrence corresponding to the item may be determined as the descriptive word corresponding to the item. Alternatively, each candidate word with a probability of occurrence corresponding to the item greater than a threshold may be determined as the descriptive word corresponding to the item. Alternatively, each candidate word may be sorted according to the size of the probability of occurrence corresponding to the item, and a preset number of candidate words may be determined as the descriptive word corresponding to the item.

[0062] It should be noted that the above examples are merely illustrative and cannot be used as limitations on the method of determining the descriptive words included in each candidate word corresponding to each item in the embodiments of the present disclosure.

[0063] It can be understood that in the embodiments of the present disclosure, by automatically mining the conversation log, the corresponding descriptive words can be determined, so that the determined descriptive words can be more in line with the user's expression, more accurate, and thus can better meet the user's needs.

[0064] In the disclosed embodiment, the target system sentence, the associated input sentence, and the object corresponding to the target system sentence contained in the dialogue log can be determined first, and then the candidate words contained in the associated input sentence can be determined according to the matching degree between the associated input sentence and each descriptive word in the preset descriptive word library, and then the corresponding relationship between each object and the associated input sentence and the candidate words contained in the associated input sentence can be determined, and then the descriptive words contained in each candidate word corresponding to each object can be determined according to the occurrence probability of the candidate word corresponding to each object. Thus, by automatically extracting and processing the dialogue log, the corresponding descriptive words can be generated, the generation efficiency of the descriptive words is higher, and the obtained descriptive words are more accurate and reliable.

[0065] Figure 2 A flowchart of a method for determining an item description word provided by an embodiment of the present disclosure is shown in FIG. Figure 2 As shown, the method for determining the item description word may include the following steps:

[0066] Step 201, determining the target system statement, the associated input statement, and the object corresponding to the target system statement contained in the dialogue log.

[0067] Step 202: segment the associated input sentence to determine each segmented word contained in the associated input sentence.

[0068] For example, the associated input sentences are "Recommend a mobile phone for the elderly", "around 2000", "a bit rough, not easy to fall off", and by performing word segmentation processing on the associated input sentences, it can be determined that the segmented words contained therein are: mobile phone for the elderly, around 2000, rough, not easy to fall off, etc., and the present disclosure does not limit this.

[0069] It is understandable that in the embodiments of the present disclosure, any desirable manner may be used to perform word segmentation processing on the associated input sentences, and the present disclosure does not limit this.

[0070] Step 203: When the matching degree between any segmented word and each description word is less than a first threshold, determine any segmented word as a candidate word.

[0071] The first threshold may be a threshold set in advance, such as 0.15, 0.2, etc., which is not limited in the present disclosure.

[0072] For example, the first threshold is 0.25. If the matching degree between any segment word A and each description word is less than the first threshold 0.25, then the segment word A can be determined as a candidate word, and the present disclosure does not limit this.

[0073] It is understandable that in the embodiments of the present disclosure, any desirable method may be used to determine the matching degree between any segmented word and each description word, and the present disclosure does not limit this.

[0074] Step 204 , determining the correspondence between each item and the candidate word according to the correspondence between each item and the associated input sentence and the candidate words included in the associated input sentence.

[0075] Step 205 : determining the first occurrence probability of each candidate word in the conversation log and the second occurrence probability in the input sentence associated with each item.

[0076] For example, there are 100 participles in the conversation log, and candidate word B appears 20 times, then the first occurrence probability corresponding to candidate word B can be 0.2; candidate word B appears 2 times in the input sentence associated with item 1, and there are 5 participles in the input sentence associated with item 1, then the second probability corresponding to candidate word B is 0.4.

[0077] It should be noted that the above examples are merely illustrative and cannot be used as limitations on the first occurrence probability, the second occurrence probability, etc. in the embodiments of the present disclosure.

[0078] Step 206: Determine the description words included in each candidate word corresponding to each item according to the first occurrence probability and / or the second occurrence probability.

[0079] Optionally, the first occurrence probability and the second occurrence probability may be added, and then the candidate word with the largest result may be determined as the descriptive word contained in the item corresponding to the candidate word. Alternatively, the candidate word with a second occurrence probability greater than the first occurrence probability may be determined as the descriptive word contained in the item corresponding to the candidate word.

[0080] Alternatively, the candidate word with the highest second occurrence probability may be determined as the descriptive word contained in the article corresponding to the candidate word. Alternatively, the candidate word with the lowest first occurrence probability may be determined as the descriptive word contained in the article corresponding to the candidate word.

[0081] It should be noted that the above examples are merely illustrative and cannot be used as limitations on the method of determining the descriptive words included in each candidate word corresponding to each item in the embodiments of the present disclosure.

[0082] Optionally, in the actual implementation process, after determining the descriptive word for each item, the second occurrence probability of the descriptive word for each item in the associated input sentence may be determined first, and then the weight value of the descriptive word for each item may be determined based on the second occurrence probability.

[0083] It is understandable that the description word may be a part of the candidate word, so the second occurrence probability of the corresponding description word may be directly determined based on the second occurrence probability of the candidate word in the input sentence associated with each item.

[0084] It can be understood that, the greater the second occurrence probability, the greater the weight value of the description word; correspondingly, the smaller the second occurrence probability, the smaller the weight value of the description word.

[0085] Therefore, in the embodiment of the present disclosure, after determining the descriptive words, the weight value of the descriptive words of each item can also be determined according to the second occurrence probability corresponding to the descriptive words. Therefore, when the determination device recommends items, it can fully consider the weights corresponding to each descriptive word, and then recommend items based on each weight, which can make the recommended items more reliable and accurate, and meet the needs of users.

[0086] In the disclosed embodiment, the target system sentence, the associated input sentence, and the object corresponding to the target system sentence contained in the dialogue log can be determined first, and then the associated input sentence can be segmented to determine each segmented word contained in the associated input sentence. When the matching degree between any segmented word and each descriptive word is less than the first threshold, any segmented word is determined as a candidate word. Then, the corresponding relationship between each object and the associated input sentence and the candidate words contained in the associated input sentence can be determined. Then, the first occurrence probability of each candidate word in the dialogue log and the second occurrence probability in the input sentence associated with each object can be determined first, and then the descriptive words contained in each candidate word corresponding to each object can be determined according to the first occurrence probability and / or the second occurrence probability. Thus, by automatically extracting and processing the dialogue log, the corresponding descriptive words can be generated, the generation efficiency of the descriptive words is higher, and the obtained descriptive words are more accurate and reliable.

[0087] Figure 3 A flowchart of a method for determining an item description word provided by an embodiment of the present disclosure is shown in FIG. Figure 3 As shown, the method for determining the item description word may include the following steps:

[0088] Step 301, determining the target system statement, the associated input statement, and the object corresponding to the target system statement contained in the dialogue log.

[0089] Step 302: Determine the candidate words included in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in the preset descriptive word library.

[0090] Step 303: Determine the correspondence between each item and the candidate word according to the correspondence between each item and the associated input sentence and the candidate words included in the associated input sentence.

[0091] Step 304 : determining the first occurrence probability of each candidate word in the conversation log and the second occurrence probability in the input sentence associated with each item.

[0092] It should be noted that the specific content and implementation of steps 301 to 304 can refer to the description of other embodiments of the present disclosure and will not be repeated here.

[0093] Step 305: Determine the first Gaussian distribution value of each candidate word in the conversation log according to each first occurrence probability.

[0094] For example, the second occurrence probability of candidate word ω in the conversation log is ω total , the mean probability of occurrence of all candidate words in the conversation log is μ total , the variance is σ total , then the first Gaussian distribution value of the candidate word ω in the conversation log can be determined according to the form shown in the following formula (1):

[0095]

[0096] Step 306: Determine the second Gaussian distribution value of the candidate word corresponding to each item in the associated input sentence according to each second occurrence probability.

[0097] For example, the probability of candidate word ω appearing in the associated input sentence is ω skuid , the mean probability of all candidate words appearing in the current item is μ skuid , the variance is σ skuid , then the second Gaussian distribution value of the candidate word ω in the associated input sentence can be determined according to the form shown in the following formula (2):

[0098]

[0099] Step 307 : Determine the description words included in each candidate word corresponding to each item according to the first occurrence probability, the second occurrence probability, the first Gaussian distribution value, and the second Gaussian distribution value.

[0100] Among them, the ratio between the second Gaussian distribution and the first Gaussian distribution can be recorded as:

[0101] score(ω)=f skuid (ω) / f total (ω) (3)

[0102] Optionally, you can first determine each candidate word whose second occurrence probability is greater than the first occurrence probability, then determine the score(ω) value corresponding to each candidate word whose second occurrence probability is greater than the first occurrence probability, arrange the score(ω) values ​​in order of size, and determine the first preset number of candidate words as the description words corresponding to the item.

[0103] Optionally, any candidate word can be determined as a description word of any item when the second occurrence probability corresponding to any candidate word corresponding to any item is greater than the first occurrence probability, and the ratio between the second Gaussian distribution value and the first Gaussian distribution value is greater than a second threshold.

[0104] The second threshold may be set in advance, such as 1.5, 1.7, etc., and the present disclosure does not limit this.

[0105] For example, the second threshold is 1.3, the second occurrence probability of any candidate word a corresponding to any item A is 0.8, the first occurrence probability is 0.4, the second probability is greater than the first probability, and the ratio of the second Gaussian distribution value to the first Gaussian distribution value is greater than the second threshold 1.3, then it can be determined that any candidate word a is a description word of any item A.

[0106] Optionally, when the second occurrence probability corresponding to any candidate word corresponding to any item is greater than the first occurrence probability, the candidate word corresponding to the largest score (ω) may be determined as the description word corresponding to the item.

[0107] It should be noted that the above examples are merely illustrative and cannot be used as limitations on the method of determining the descriptive words included in each candidate word corresponding to each item in the embodiments of the present disclosure.

[0108] In the disclosed embodiment, the target system sentence, the associated input sentence, and the object corresponding to the target system sentence contained in the dialogue log can be determined first, and then the candidate words contained in the associated input sentence can be determined according to the matching degree between the associated input sentence and each descriptive word in the preset descriptive word library, and then the corresponding relationship between each object and the associated input sentence and the candidate words contained in the associated input sentence can be determined, and then the first occurrence probability of each candidate word in the dialogue log and the second occurrence probability in the input sentence associated with each object can be determined, and the first Gaussian distribution value of each candidate word in the dialogue log and the second Gaussian distribution value of the candidate word corresponding to each object in the associated input sentence can be determined, and then the descriptive words contained in each candidate word corresponding to each object can be determined according to the first occurrence probability, the second occurrence probability, the first Gaussian distribution value, and the second Gaussian distribution value. Thus, by automatically extracting and processing the dialogue log, the corresponding descriptive words can be generated, the generation efficiency of the descriptive words is higher, and the obtained descriptive words are more accurate and reliable.

[0109] Figure 4 A flowchart of a method for determining an item description word provided by an embodiment of the present disclosure is shown in FIG. Figure 4 As shown, the method for determining the item description word may include the following steps:

[0110] Step 401: Receive a query statement.

[0111] It is understandable that the determination device can determine to receive the query statement after detecting that the user triggers the "Send" control; or can determine to receive the query statement after detecting that the user inputs relevant content, etc., and the present disclosure does not limit this.

[0112] Step 402: Process the query statement based on the description word library to determine the target description word contained in the query statement.

[0113] The description word library may include description words, which may be any words that describe item-related information, such as applicable age, season, feel, item name, model, etc., which is not limited in the present disclosure.

[0114] The query statement may be segmented to obtain the word segments corresponding to the query statement, and then the word segments may be matched with the description words in the description word library. The description words that meet the matching conditions are the target description words contained in the query statement.

[0115] For example, the descriptive word with the highest matching degree is used as the target descriptive word included in the query statement; or, each descriptive word with a matching degree greater than a threshold can be used as the target descriptive word included in the query statement, and the present disclosure does not limit this.

[0116] It should be noted that in the embodiments of the present disclosure, any desirable method may be used to determine the matching degree between the segmented words and the description words in the description word library, and the present disclosure does not limit this.

[0117] Step 403: Determine candidate items associated with the target descriptive word according to the preset association relationship between the descriptive word and the item.

[0118] For example, if the target descriptive words are "senior mobile phone, 1,000 yuan", then the items associated with "senior mobile phone" and "1,000 yuan" can be traversed in the preset association relationship between descriptive words and items. For example, if the items associated with "senior mobile phone" and "1,000 yuan" are: mobile phone 1, mobile phone 2, then the candidate items associated with the target descriptive words can be determined to be: mobile phone 1, mobile phone 2.

[0119] It should be noted that the above examples are merely illustrative and cannot be used as limitations on the target descriptors and associated candidate items in the embodiments of the present disclosure.

[0120] Step 404 , determining the target item according to the weight of the target descriptor in each candidate item descriptor.

[0121] For example, the weight of the target description word "mobile phone for the elderly" in "mobile phone 1" is 0.5, the weight of "1,000 yuan" in "mobile phone 1" is 0.2, and the sum of their weights is 0.7; and the weight of the target description word "mobile phone for the elderly" in candidate item 2 of "mobile phone 2" is 0.3, and its weight in "mobile phone 2" is 0.1, and the sum of their weights is 0.4, so "mobile phone 1" can be determined as the target item.

[0122] It should be noted that the above examples are merely illustrative and cannot be used as limitations on the method of determining the target object in the embodiments of the present disclosure.

[0123] In the disclosed embodiment, when determining the target item, the weight of each target descriptor in the candidate item descriptors is fully considered, so that based on each weight, the determined target item is more accurate, reliable, and more in line with user needs.

[0124] Step 405, return the target item.

[0125] It is understandable that after determining the target object, the determination device may send the target object to the user so that the user can get feedback in time, thereby giving the user a good experience.

[0126] In the disclosed embodiment, a query statement may be received first, and then the query statement may be processed based on the description word library to determine the target descriptive word contained in the query statement, and then the candidate items associated with the target descriptive word may be determined based on the association relationship between the preset descriptive word and the item, and then the target item may be determined based on the weight of the target descriptive word in each candidate item descriptive word, and then the target item may be returned. Thus, when recommending items, the weight corresponding to the target descriptive word is fully considered, so that the recommended items can be more accurate, more in line with user needs, and can give users a good sense of use.

[0127] In order to implement the above embodiment, the present disclosure also proposes a device for determining an item description word.

[0128] Figure 5 A schematic diagram of the structure of a device for determining an item description word provided in an embodiment of the present disclosure.

[0129] like Figure 5 As shown, the device 100 for determining an item description word may include: a first determining module 110 , a second determining module 120 , a third determining module 130 and a fourth determining module 140 .

[0130] The first determination module 110 is used to determine the target system statement, the associated input statement, and the item corresponding to the target system statement contained in the conversation log.

[0131] A second determination module 120, configured to determine candidate words included in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in a preset descriptive word library;

[0132] A third determination module 130, configured to determine a correspondence between each item and a candidate word according to the correspondence between each item and the associated input sentence and the candidate words included in the associated input sentence;

[0133] The fourth determination module 140 is used to determine the description words included in each candidate word corresponding to each of the items according to the occurrence probability of the candidate word corresponding to each of the items.

[0134] Optionally, the first determining module 110 is specifically configured to:

[0135] Determine the system statement in the conversation log that points to the page where the item is located as the target system statement;

[0136] Determine the item according to the description information of the page where the item pointed to by the target system statement is located;

[0137] A preset number of input sentences whose reception times are before the target system sentence are determined as the associated input sentences.

[0138] Optionally, the second determining module 120 is specifically configured to:

[0139] Segmenting the associated input sentence to determine each segmented word contained in the associated input sentence;

[0140] When the matching degree between any participle and each of the description words is less than a first threshold, the any participle is determined to be a candidate word.

[0141] Optionally, the fourth determining module 140 is specifically configured to:

[0142] Determine a first occurrence probability of each candidate word in the conversation log and a second occurrence probability in an input sentence associated with each item;

[0143] Describe words included in each candidate word corresponding to each of the items according to the first occurrence probability and / or the second occurrence probability.

[0144] Optionally, the fourth determining module 140 includes:

[0145] A first determination unit, configured to determine a first occurrence probability of each candidate word in the conversation log and a second occurrence probability in an input sentence associated with each item;

[0146] a second determining unit, configured to determine a first Gaussian distribution value of each of the candidate words in the conversation log according to each of the first occurrence probabilities;

[0147] a third determining unit, configured to determine, according to each of the second occurrence probabilities, a second Gaussian distribution value of the candidate word corresponding to each of the items in the associated input sentence;

[0148] The fourth determining unit is used to determine the description words included in each candidate word corresponding to each of the items according to the first occurrence probability, the second occurrence probability, the first Gaussian distribution value and the second Gaussian distribution value.

[0149] Optionally, the fourth determining unit is specifically configured to:

[0150] In response to the second occurrence probability corresponding to any candidate word corresponding to any item being greater than the first occurrence probability, and the ratio between the second Gaussian distribution value and the first Gaussian distribution value being greater than a second threshold, the any candidate word is determined to be a description word of the any item.

[0151] Optionally, the first determining module 110 is further configured to:

[0152] Determining a second occurrence probability of each of the item's description words in the associated input sentence;

[0153] A weight value of the description word of each of the items is determined according to the second occurrence probability.

[0154] Optionally, the above device further includes:

[0155] The receiving module is used to receive query statements.

[0156] The processing module is used to process the query statement based on the description word library to determine the target description word contained in the query statement.

[0157] The fifth determination module is used to determine the candidate items associated with the target descriptive word according to the preset association relationship between the descriptive word and the item.

[0158] The sixth determination module is used to determine the target item according to the weight of the target descriptor in each of the candidate item descriptors.

[0159] The return module is used to return the target item.

[0160] The functions and specific implementation principles of the above modules in the embodiments of the present disclosure can be referred to the above method embodiments, and will not be repeated here.

[0161] The device for determining the item description words of the disclosed embodiment can first determine the target system sentence, the associated input sentence, and the item corresponding to the target system sentence contained in the dialogue log, and then determine the candidate words contained in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in the preset descriptive word library, and then determine the corresponding relationship between each item and the associated input sentence and the candidate words contained in the associated input sentence, and then determine the descriptive words contained in each candidate word corresponding to each item according to the occurrence probability of the candidate word corresponding to each item. Thus, by automatically extracting and processing the dialogue log, the corresponding descriptive words can be generated, the generation efficiency of the descriptive words is higher, and the obtained descriptive words are more accurate and reliable.

[0162] In order to implement the above embodiments, the present disclosure further proposes a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for determining the item description words proposed in the above embodiments of the present disclosure is implemented.

[0163] In order to implement the above embodiments, the present disclosure further proposes a non-temporary computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for determining the item descriptor proposed in the above embodiments of the present disclosure.

[0164] In order to implement the above embodiments, the present disclosure further proposes a computer program product. When the instruction processor in the computer program product is executed, the method for determining the item descriptor proposed in the above embodiments of the present disclosure is executed.

[0165] Figure 6 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. Figure 6 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0166] like Figure 6 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0167] The bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnection (PCI) bus.

[0168] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0169] The memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6 Not shown in the figure, a disk drive for reading and writing a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing a removable non-volatile optical disk (e.g., a compact disc read only memory (CD-ROM), a digital versatile disc read only memory (DVD-ROM), or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the various embodiments of the present disclosure.

[0170] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in the memory 28, such program modules 42 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 42 generally perform the functions and / or methods of the embodiments described in the present disclosure.

[0171] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed via an input / output (I / O) interface 22. In addition, the computer device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the computer device 12 via a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0172] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the methods mentioned in the above embodiments.

[0173] The technical solution disclosed in the present invention can first determine the target system sentence, the associated input sentence, and the object corresponding to the target system sentence contained in the dialogue log, and then determine the candidate words contained in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in the preset descriptive word library, and then determine the corresponding relationship between each object and the associated input sentence and the candidate words contained in the associated input sentence, and then determine the descriptive words contained in each candidate word corresponding to each object according to the occurrence probability of the candidate word corresponding to each object. Thus, by automatically extracting and processing the dialogue log, the corresponding descriptive words can be generated, the generation efficiency of the descriptive words is higher, and the obtained descriptive words are more accurate and reliable.

[0174] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0175] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0176] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present disclosure belong.

[0177] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0178] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0179] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0180] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0181] The storage medium mentioned above may be a read-only memory, a disk or an optical disk, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present disclosure. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present disclosure.

Claims

1. A method for determining an item description word, It is characterized in that include: Determine a target system statement, an associated input statement, and an item corresponding to the target system statement contained in the conversation log, wherein the target system statement is a system statement in the conversation log used to point to a page where an item is located, the item is determined based on description information of the page where the item pointed to by the target system statement is located, and the associated input statement is a preset number of input statements whose reception time is located before the target system statement; Determining candidate words included in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in a preset descriptive word library; Determining the correspondence between each item and the candidate word according to the correspondence between each item and the associated input sentence and the candidate words included in the associated input sentence; According to the occurrence probability of the candidate words corresponding to each of the items, the descriptive words included in each of the candidate words corresponding to each of the items are determined.

2. The method according to claim 1, It is characterized in that The determining of the candidate words contained in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in a preset descriptive word library includes: Segmenting the associated input sentence to determine each segmented word contained in the associated input sentence; When the matching degree between any participle and each of the description words is less than a first threshold, the any participle is determined to be a candidate word.

3. The method according to claim 1, It is characterized in that The step of determining the descriptive words included in each candidate word corresponding to each item according to the occurrence probability of the candidate word corresponding to each item includes: Determine a first occurrence probability of each candidate word in the conversation log and a second occurrence probability in an input sentence associated with each item; Describe words included in each candidate word corresponding to each of the items according to the first occurrence probability and / or the second occurrence probability.

4. The method according to claim 1, It is characterized in that The step of determining the descriptive words included in each candidate word corresponding to each item according to the occurrence probability of the candidate word corresponding to each item includes: Determine a first occurrence probability of each candidate word in the conversation log and a second occurrence probability in an input sentence associated with each item; Determining a first Gaussian distribution value of each candidate word in the conversation log according to each of the first occurrence probabilities; Determining, according to each of the second occurrence probabilities, a second Gaussian distribution value of the candidate word corresponding to each of the items in the associated input sentence; Describe words included in each candidate word corresponding to each of the items according to the first occurrence probability, the second occurrence probability, the first Gaussian distribution value, and the second Gaussian distribution value.

5. The method according to claim 4, It is characterized in that The step of determining the description words included in each candidate word corresponding to each of the items according to the first occurrence probability, the second occurrence probability, the first Gaussian distribution value, and the second Gaussian distribution value includes: In response to the second occurrence probability corresponding to any candidate word corresponding to any item being greater than the first occurrence probability, and the ratio between the second Gaussian distribution value and the first Gaussian distribution value being greater than a second threshold, the any candidate word is determined to be a description word of the any item.

6. The method according to any one of claims 1 to 5, It is characterized in that Also includes: Determining a second occurrence probability of each of the item's description words in the associated input sentence; A weight value of the description word of each of the items is determined according to the second occurrence probability.

7. The method according to claim 6, It is characterized in that After determining the weight value of each descriptive word of the item, the method further includes: Receive query statements; Processing the query statement based on a description word library to determine a target description word included in the query statement; Determining candidate items associated with the target descriptive word based on the association relationship between the preset descriptive word and the item; Determining a target item according to a weight of the target descriptor in each of the candidate item descriptors; Returns the target item.

8. A device for determining a description word of an item, It is characterized in that include: A first determination module is used to determine a target system statement, an associated input statement, and an item corresponding to the target system statement contained in a conversation log, wherein the target system statement is a system statement in the conversation log that points to a page where an item is located, the item is determined based on description information of the page where the item pointed to by the target system statement is located, and the associated input statement is a preset number of input statements whose reception time is located before the target system statement; A second determination module, configured to determine the candidate words included in the associated input sentence according to the matching degree between the associated input sentence and each descriptive word in a preset descriptive word library; A third determination module, configured to determine a correspondence between each of the items and the candidate words according to the correspondence between each item and the associated input sentence and the candidate words included in the associated input sentence; The fourth determination module is used to determine the descriptive words included in each candidate word corresponding to each of the items according to the occurrence probability of the candidate word corresponding to each of the items.

9. The device as claimed in claim 8, It is characterized in that The second determining module is specifically used to: Segmenting the associated input sentence to determine each segmented word contained in the associated input sentence; When the matching degree between any participle and each of the description words is less than a first threshold, the any participle is determined to be a candidate word.

10. The device according to claim 8, It is characterized in that The fourth determining module is specifically used for: Determine a first occurrence probability of each candidate word in the conversation log and a second occurrence probability in an input sentence associated with each item; Describe words included in each candidate word corresponding to each of the items according to the first occurrence probability and / or the second occurrence probability.

11. The device according to claim 8, It is characterized in that The fourth determining module includes: A first determination unit, configured to determine a first occurrence probability of each candidate word in the conversation log and a second occurrence probability in an input sentence associated with each item; a second determining unit, configured to determine a first Gaussian distribution value of each of the candidate words in the conversation log according to each of the first occurrence probabilities; a third determining unit, configured to determine, according to each of the second occurrence probabilities, a second Gaussian distribution value of the candidate word corresponding to each of the items in the associated input sentence; The fourth determining unit is used to determine the description words included in each candidate word corresponding to each of the items according to the first occurrence probability, the second occurrence probability, the first Gaussian distribution value and the second Gaussian distribution value.

12. The device according to claim 11, It is characterized in that The fourth determining unit is specifically configured to: In response to the second occurrence probability corresponding to any candidate word corresponding to any item being greater than the first occurrence probability, and the ratio between the second Gaussian distribution value and the first Gaussian distribution value being greater than a second threshold, the any candidate word is determined to be a description word of the any item.

13. The device according to any one of claims 8 to 12, It is characterized in that The first determining module is further used for: Determining a second occurrence probability of each of the item's description words in the associated input sentence; A weight value of the description word of each of the items is determined according to the second occurrence probability.

14. The device according to claim 13, It is characterized in that Also includes: A receiving module, used for receiving a query statement; A processing module, configured to process the query statement based on a description word library to determine a target description word included in the query statement; A fifth determination module, configured to determine candidate items associated with the target descriptive word according to a preset association relationship between the descriptive word and the item; A sixth determination module, configured to determine a target item according to a weight of the target descriptor in each of the candidate item descriptors; The return module is used to return the target item.

15. A computer device, It is characterized in that The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for determining the item description word according to any one of claims 1 to 7 is implemented.

16. A computer-readable storage medium storing a computer program, It is characterized in that When the computer program is executed by a processor, the method for determining an item descriptor as claimed in any one of claims 1 to 7 is implemented.

17. A computer program product, It is characterized in that The invention comprises a computer program, which, when executed by a processor, implements the method for determining an item descriptor according to any one of claims 1 to 7.

Images