A method and device for sorting associative words

Abstract

The present invention discloses a method and device for sorting associative words, and relates to the field of computer technology. A specific implementation of the method includes: obtaining stroke information to be associated, determining the heat value and business data value of each associative word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character; determining the stroke similarity between the stroke information and each associative word; and sorting each associative word corresponding to the stroke information according to the heat value, business data value and stroke similarity. This implementation can solve the problem that associative words corresponding to strokes cannot be sorted.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for sorting associative words. Background Art

[0002] The calculation of associative words is widely used in business fields such as search engines and input methods. After a user enters characters, the priority of associative words similar to or related to the input characters can be calculated, and the associative words can be sorted according to the priority to facilitate user selection.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art: the existing associative word calculation model generates all stroke-associative word pairs indiscriminately and cannot prioritize the associative words corresponding to the strokes; and generating all stroke-associative word pairs will lead to a waste of resources. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method and device for sorting associative words, so as to solve the technical problem that associative words corresponding to strokes cannot be sorted.

[0005] To achieve the above object, according to one aspect of an embodiment of the present invention, a method for sorting associative words is provided, comprising:

[0006] Acquire stroke information to be associated, and determine the heat value and business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character;

[0007] Determining the stroke similarity between the stroke information and the strokes of each of the associated words;

[0008] According to the heat value, the business data value and the stroke similarity, the associative words corresponding to the stroke information are sorted.

[0009] Optionally, determining the heat value of each associative word corresponding to the stroke information includes:

[0010] Calculating the Bayesian mean and the Newton cooling value respectively according to the number of searches for each associated word corresponding to the stroke information;

[0011] The heat value of each associative word corresponding to the stroke information is calculated according to the Bayesian mean value and the Newton cooling value.

[0012] Optionally, determining the heat value of each associative word corresponding to the stroke information includes:

[0013] Calculate the Bayesian average of each associative word by using the Bayesian formula, the Bayesian formula is B(wi)=(Atp(wi))+Tp(avg)*R(avg)) / (TpSum(wi)+Tp(avg)), i=0,1,2,…,m, wherein m is the total number of all associative words corresponding to the stroke information, B(wi) is the Bayesian average of the ith associative word, Atp(wi) is the current search number of the ith associative word, TpSum(wi) is the sum of the historical search number and the current search number of the ith associative word, Tp(avg) is the average of the total search number of all associative words, and R(avg) is the average of Atp(wi) / TpSum(wi) of all associative words;

[0014] Calculating the Newton cooling value of each associative word by the Newton cooling formula, the Newton cooling formula is C(wi)=log(Atp(wi)+1 / Btp(wi)+1), wherein C(wi) is the Newton cooling value of the i-th associative word, and Btp(wi) is the number of historical searches of the i-th associative word;

[0015] The heat value of the i-th associative word is calculated according to the formula H(wi)=X*B(wi)+Y*C(wi), where H(wi) is the heat value of the i-th associative word, 0<X<1, 0<Y<1, and X+Y=1.

[0016] Optionally, determining the business data value of each associative word corresponding to the stroke information includes:

[0017] Normalizing the sales data, click-through rate and click-through conversion rate of each associated word respectively;

[0018] The business data value of each of the association words is calculated based on the normalized sales data, click-through rate and click-through conversion rate, and their respective weights.

[0019] Optionally, determining the stroke similarity between the stroke information and the strokes of each of the associated words includes:

[0020] The stroke similarity of each associated word is calculated according to the number of strokes included in the stroke information and the total number of strokes included in each associated word.

[0021] Optionally, before obtaining the stroke information to be associated and determining the heat value and business data value of each associated word corresponding to the stroke information, the method further includes:

[0022] According to the search data of the associated words, filter the associated words with the highest search data;

[0023] Stroke information corresponding to the associative word is constructed, and a mapping relationship between the associative word and each stroke information corresponding thereto is obtained.

[0024] In addition, according to another aspect of the embodiment of the present invention, a device for sorting associative words is provided, comprising:

[0025] A first determination module is used to obtain stroke information to be associated, and determine the heat value and business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character;

[0026] A second determination module is used to determine the stroke similarity between the stroke information and the strokes of each associated word;

[0027] The sorting module is used to sort the associated words corresponding to the stroke information according to the heat value, the business data value and the stroke similarity.

[0028] Optionally, determining the heat value of each associative word corresponding to the stroke information includes:

[0029] Calculating the Bayesian mean and the Newton cooling value respectively according to the number of searches for each associated word corresponding to the stroke information;

[0030] The heat value of each associative word corresponding to the stroke information is calculated according to the Bayesian mean value and the Newton cooling value.

[0031] Optionally, determining the heat value of each associative word corresponding to the stroke information includes:

[0032] Calculate the Bayesian average of each associative word by using the Bayesian formula, the Bayesian formula is B(wi)=(Atp(wi))+Tp(avg)*R(avg)) / (TpSum(wi)+Tp(avg)), i=0,1,2,…,m, wherein m is the total number of all associative words corresponding to the stroke information, B(wi) is the Bayesian average of the ith associative word, Atp(wi) is the current search number of the ith associative word, TpSum(wi) is the sum of the historical search number and the current search number of the ith associative word, Tp(avg) is the average of the total search number of all associative words, and R(avg) is the average of Atp(wi) / TpSum(wi) of all associative words;

[0033] Calculating the Newton cooling value of each associative word by the Newton cooling formula, the Newton cooling formula is C(wi)=log(Atp(wi)+1 / Btp(wi)+1), wherein C(wi) is the Newton cooling value of the i-th associative word, and Btp(wi) is the number of historical searches of the i-th associative word;

[0034] The heat value of the i-th associative word is calculated according to the formula H(wi)=X*B(wi)+Y*C(wi), where H(wi) is the heat value of the i-th associative word, 0<X<1, 0<Y<1, and X+Y=1.

[0035] Optionally, determining the business data value of each associative word corresponding to the stroke information includes:

[0036] Normalizing the sales data, click-through rate and click-through conversion rate of each associated word respectively;

[0037] The business data value of each of the association words is calculated based on the normalized sales data, click-through rate and click-through conversion rate, and their respective weights.

[0038] Optionally, the second determining module is used to:

[0039] The stroke similarity of each associated word is calculated according to the number of strokes included in the stroke information and the total number of strokes included in each associated word.

[0040] Optionally, the device further comprises:

[0041] The construction module is used to filter the associative words with the highest search data according to the search data of the associative words; construct the stroke information corresponding to the associative words, and obtain the mapping relationship between the associative words and the stroke information corresponding to them.

[0042] According to another aspect of an embodiment of the present invention, there is provided a method for information search, comprising:

[0043] receiving a search request from a user, wherein the search request carries stroke information to be associated;

[0044] Parsing the search request to obtain stroke information to be associated;

[0045] Determining a heat value and a business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character;

[0046] Determining the stroke similarity between the stroke information and the strokes of each of the associated words;

[0047] sorting the associated words corresponding to the stroke information according to the heat value, the business data value and the stroke similarity;

[0048] The sorting results of each associated word corresponding to the stroke information are pushed to the user.

[0049] Optionally, determining the heat value of each associative word corresponding to the stroke information includes:

[0050] Calculating the Bayesian mean and the Newton cooling value respectively according to the number of searches for each associated word corresponding to the stroke information;

[0051] The heat value of each associative word corresponding to the stroke information is calculated according to the Bayesian mean value and the Newton cooling value.

[0052] Optionally, calculating the heat value of each associative word corresponding to the stroke information includes:

[0053] The Bayesian average of each associative word is calculated by the Bayesian formula, where the Bayesian formula is B(wi)=(Atp(wi))+Tp(avg)*R(avg)) / (TpSum(wi)+Tp(avg)), i=0,1,2,…,m;

[0054] Wherein, m is the total number of all associated words corresponding to the stroke information, B(wi) is the Bayesian mean of the i-th associated word, Atp(wi) is the current search count of the i-th associated word, TpSum(wi) is the sum of the historical search count and the current search count of the i-th associated word, Tp(avg) is the average of the sum of the search counts of all associated words, and R(avg) is the average of Atp(wi) / TpSum(wi) of all associated words;

[0055] The Newton cooling value of each associative word is calculated by the Newton cooling formula, where the Newton cooling formula is C(wi)=log(Atp(wi)+1 / Btp(wi)+1);

[0056] Where C(wi) is the Newton cooling value of the i-th associative word, and Btp(wi) is the number of historical searches for the i-th associative word;

[0057] Calculate the heat value of the i-th associative word according to the formula H(wi)=X*B(wi)+Y*C(wi);

[0058] Among them, H(wi) is the heat value of the i-th associative word, 0<X<1, 0<Y<1, X+Y=1.

[0059] Optionally, determining the business data value of each associative word corresponding to the stroke information includes:

[0060] Normalizing the sales data, click-through rate and click-through conversion rate of each associated word respectively;

[0061] The business data value of each of the association words is calculated based on the normalized sales data, click-through rate and click-through conversion rate, and their respective weights.

[0062] Optionally, determining the stroke similarity between the stroke information and the strokes of each of the associated words includes:

[0063] The stroke similarity of each associated word is calculated according to the number of strokes included in the stroke information and the total number of strokes included in each associated word.

[0064] Optionally, before obtaining the stroke information to be associated and determining the heat value and business data value of each associated word corresponding to the stroke information, the method further includes:

[0065] According to the search data of the associated words, filter the associated words with the highest search data;

[0066] Stroke information corresponding to the associative word is constructed, and a mapping relationship between the associative word and each stroke information corresponding thereto is obtained.

[0067] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, including:

[0068] one or more processors;

[0069] a storage device for storing one or more programs,

[0070] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the above embodiments.

[0071] According to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0072] An embodiment of the above invention has the following advantages or beneficial effects: because the technical means of sorting the associated words corresponding to the stroke information to be associated according to the heat value, the business data value and the stroke similarity is adopted, the technical problem of being unable to sort the associated words corresponding to the strokes is overcome, and the associated words corresponding to the stroke information to be associated are sorted comprehensively according to the heat value, the business data value and the stroke similarity, thereby avoiding the indiscriminate generation of all stroke-associative word pairs. Therefore, the method provided by the embodiment of the present invention sorts the associated words according to the characteristics of the associated words themselves, thereby generating stroke-associative word pairs differently, which can avoid waste of resources and improve the speed of processing data.

[0073] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0075] Figure 1is a schematic diagram of the main process of the method for sorting associative words according to an embodiment of the present invention;

[0076] Figure 2 is a schematic diagram of the main process of a method for sorting associative words according to a reference embodiment of the present invention;

[0077] Figure 3 is a schematic diagram of main modules of a device for sorting associative words according to an embodiment of the present invention;

[0078] Figure 4 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0079] Figure 5 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0080] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0081] Figure 1 Schematic diagram of the main process of the method for sorting associative words according to an embodiment of the present invention. As an embodiment of the present invention, Figure 1 As shown, the method for sorting the associative words may include:

[0082] Step 101, obtaining stroke information to be associated, and determining the heat value and business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character.

[0083] In this step, the stroke information input by the user, i.e., the stroke information to be associated, is obtained, and the associated words corresponding to the stroke information to be associated are determined, and finally the heat value and business data value of each associated word are determined. It should be noted that the stroke information can be at least one stroke, at least one character, or a combination of strokes and characters.

[0084] For example, if the stroke information input by the user is "穴", according to the stroke input method, it can be known that this stroke information is composed of the five basic strokes "乀丿乛丿乀" (捺, 撇, 折, 撇, 捺). Based on these five basic strokes, all the associated words corresponding to this stroke information are determined (such as air conditioner, air, air freshener, space, knack, etc.), and then the popularity value and business data value of each associated word are determined. It should be noted that the embodiments of the present invention are not limited to Chinese strokes and can also be applicable to Japanese strokes.

[0085] As another embodiment of the present invention, the step of determining the popularity value of each associated word corresponding to the stroke information may include: calculating the Bayesian average value and the Newton cooling value respectively according to the search times of each associated word corresponding to the stroke information; calculating the popularity value of each associated word corresponding to the stroke information based on the Bayesian average value and the Newton cooling value to ensure the effectiveness and reliability of the popularity value. Specifically, in this step, the Bayesian average value of each associated word can be calculated first through the Bayesian formula, and then the Newton cooling value of each associated word can be calculated through the Newton cooling formula. Finally, based on the Bayesian average value and the Newton cooling value, the popularity value of each associated word is obtained.

[0086] As another embodiment of the present invention, the calculation step of the popularity value may include:

[0087] First, calculate the Bayesian average value of each associated word through the Bayesian formula. The Bayesian formula is B(wi) = (Atp(wi) + Tp(avg) * R(avg)) / (TpSum(wi) + Tp(avg)), where i = 0, 1, 2,..., m;

[0088] where m is the total number of all associated words corresponding to the stroke information;

[0089] B(wi) is the Bayesian average value of the i-th associated word;

[0090] Atp(wi) is the current search times of the i-th associated word;

[0091] TpSum(wi) is the sum of the historical search times and the current search times of the i-th associated word;

[0092] Tp(avg) is the average value of the sum of the search times of all associated words (i.e., the sum of TpSum(wi) of all associated words / the total number of associated words);

[0093] R(avg) is the average value of Atp(wi) / TpSum(wi) of all associated words (i.e., the sum of Atp(wi) / TpSum(wi) of all associated words / the total number of associated words);

[0094] Then, the Newton cooling value of each associative word is calculated by the Newton cooling formula, and the Newton cooling formula is C(wi)=log(Atp(wi)+1 / Btp(wi)+1);

[0095] Among them, C(wi) is the Newton cooling value of the i-th associative word,

[0096] Btp(wi) is the historical search times of the i-th associative word;

[0097] Finally, the heat value of the i-th associative word is calculated according to the formula H(wi)=X*B(wi)+Y*C(wi);

[0098] Among them, H(wi) is the heat value of the i-th associative word, 0<X<1, 0<Y<1, X+Y=1.

[0099] It should be noted that X is the weight coefficient of the Bayesian average value, and Y is the weight coefficient of the Newtonian cooling value. X and Y can be determined according to the weights of the Bayesian average value and the Newtonian cooling value, or different values ​​can be set as needed. Optionally, the heat value of the i-th associative word is calculated according to the formula H(wi)=0.8*B(wi)+0.2*C(wi).

[0100] It should be pointed out that the current number of searches refers to the number of searches in the most recent short period of time, such as the last 1 hour, 5 hours, day, three days, week, etc.; the historical number of searches refers to the number of searches in the most recent longer period of time minus the current number of searches, such as the last month, two months, three months, half a year, year, etc., minus the current number of searches.

[0101] Therefore, the method provided by the embodiment of the present invention can accurately calculate the heat value of each associated word through the Bayesian formula and the Newton cooling formula, thereby providing effective and reliable basic data for subsequent steps. Moreover, using the heat value as the basis for sorting can more accurately sort the associated words corresponding to the stroke information, reducing the user's screening workload.

[0102] Optionally, according to different application scenarios, business data values ​​corresponding to different application scenarios are calculated. For example, for an e-commerce scenario, the business data values ​​may include sales data corresponding to the associated words, click-through rates corresponding to the associated words, and click-through conversion rates corresponding to the associated words to reduce the amount of data.

[0103] As another embodiment of the present invention, determining the business data value of each associative word corresponding to the stroke information includes: calculating the business data value of each associative word according to the sales data, click-through rate and click-through conversion rate of each associative word and their respective weights. Specifically, respectively obtain the sales data, click-through rate and click-through conversion rate corresponding to each associative word, and then respectively calculate the business data value of each associative word based on these data and their respective weights.

[0104] Among them, the sales data can be the sales volume corresponding to a certain associative word, such as the historical sales volume in a recent period of time. The click-through rate can refer to the ratio of the number of times a certain associative word is clicked to the number of times it is displayed on the website page. The click-through conversion rate can refer to the number of transactions corresponding to a certain associative word and the number of times the associative word is clicked.

[0105] Optionally, the business data value of each associative word is calculated based on different weight coefficients corresponding to sales data, click-through rate, and click-through conversion rate. For example, business data value = S*sales + P*click-through rate + Q*click-through conversion rate, where S is the weight coefficient of sales, P is the weight coefficient of click-through rate, and Q is the weight coefficient of click-through conversion rate. Accordingly, 0<S<1, 0<P<1, 0<Q<1, S+P+Q=1. S, P, Q can be determined based on the weights of sales data, click-through rate, and click-through conversion rate, and different values ​​can be set as needed. For example: business data value = 0.6*sales + 0.2*click-through rate + 0.2*click-through conversion rate.

[0106] As another embodiment of the present invention, determining the business data value of each associative word corresponding to the stroke information includes: first, normalizing the sales data, click-through rate and click-through conversion rate of each associative word respectively; then, calculating the business data value of each associative word according to the normalized sales data, click-through rate and click-through conversion rate, and their respective weights. In this embodiment, the sales amount is the normalized sales amount (nor_gmv), the click-through rate is the normalized click-through rate (nor_ctr), and the click-through conversion rate is the normalized click-through conversion rate (nor_cvr), so as to unify the results of the three indicators to the same scale, thereby improving the validity and reliability of the business data value.

[0107] After normalization, the values ​​of sales, click-through rate, and click-through conversion rate are all between 0 and 1. Therefore, in this embodiment, 0<S<100, 0<P<100, 0<Q<100, S+P+Q=100. For example: business data value=60*sales+20*click-through rate+20*click-through conversion rate.

[0108] Step 102: Determine the stroke similarity between the stroke information and the strokes of each associated word.

[0109] Optionally, step 102 includes: determining the stroke similarity between the stroke information and each of the associated words according to the number of strokes included in the stroke information and the total number of strokes included in each of the associated words.

[0110] Specifically, obtain the number of strokes included in the stroke information and the total number of strokes included in each of the associated words respectively, and then calculate the ratio of the number of strokes included in the stroke information to the total number of strokes included in a certain associated word respectively.

[0111] For example, stroke similarity = 100 * number of strokes included in the stroke information / total number of strokes included in the associated word. For example, for stroke information: "穴" (cave), associated word: "空" (empty), then the stroke similarity between "穴" and "空" = 100 * 5 / 8 = 62.5; for stroke information: "手一丨" (hand, one, vertical stroke), associated word: "手机" (mobile phone), then the stroke similarity between "手一丨" and "手机" = 100 * 6 / 10 = 60.

[0112] Therefore, taking the stroke similarity between the stroke information and the associated words as basic data provides effective and reliable sorting for subsequent steps, and the stroke similarity can further improve the association accuracy between the stroke information and the associated words.

[0113] It should be noted that step 101 can be executed first, followed by step 102, or step 102 can be executed first, followed by step 101, or step 101 and step 102 can be executed simultaneously. The present invention does not limit this.

[0114] Step 103, sort each of the associated words corresponding to the stroke information according to the heat value, business data value, and stroke similarity.

[0115] Optionally, calculate the total score of each of the associated words according to the weights of the heat value, business data value, and stroke similarity, and sort each of the associated words according to the total scores of each of the associated words, so as to determine the priority of each of the associated words.

[0116] Specifically, the following formula can be used: total score = U * heat value + V * business data value + W * stroke similarity, where 0 < U < 100, 0 < V < 100, 0 < W < 100, and U + V + W = 100. U, V, and W can be determined according to the weights of the heat value, business data value, and stroke similarity, and different values can be set as needed. For example: total score = 60 * heat value + 20 * business data value + 20 * stroke similarity.

[0117] It should be noted that the heat value, business data value, and stroke similarity value need to be unified to the same scale to ensure the accuracy of the total score calculated based on these three factors. Optionally, the method provided by the embodiments of the present invention can implement data calculation based on the Spark distributed memory computing framework, which can quickly process data above the T level and support fault-tolerant computing.

[0118] According to the various embodiments described above, it can be seen that the present invention solves the problem of being unable to sort the associative words corresponding to strokes by using the technical means of sorting the associative words corresponding to the stroke information to be associated according to the heat value, business data value, and stroke similarity. That is to say, the prior art generates all stroke-associative word pairs without discrimination, resulting in waste of resources. The present invention comprehensively sorts the associative words corresponding to the stroke information to be associated according to the heat value, business data value, and stroke similarity, so as to avoid generating all stroke-associative word pairs without discrimination. Therefore, the method provided by the embodiments of the present invention sorts each associative word according to the characteristics of the associative word itself (i.e., heat value, business data value, and stroke similarity), so as to generate stroke-associative word pairs differentially, which can not only avoid waste of resources but also improve the speed of processing data.

[0119] As another embodiment of the present invention, before step 101, it further includes: first, screening the associative words with the top search data according to the search data of the associative words; then, constructing the stroke information corresponding to the associative words and obtaining the mapping relationship between the associative words and their corresponding stroke information. In this embodiment, the search ranking of the associative words can be determined according to the search data of the associative words, so as to differentially construct the stroke information of the associative words according to the search ranking.

[0120] For example, based on the stroke input method (including five basic strokes, namely horizontal, vertical, left-falling, right-falling, and turning), for the word data (i.e., associative words) with the top 10% search ranking, the stroke information of all the characters in the associative words is constructed. Taking the mobile phone as an example:

[0121] 丿→mobile phone, 丿一→mobile phone, 丿一一→mobile phone, 丿一一丨→mobile phone, 手一→mobile phone, 手一丨→mobile phone, 手一丨丿→mobile phone, 手一丨丿乀→mobile phone, 手一丨丿乀丿→mobile phone, 手一丨丿乀丿乛→mobile phone. That is, 10 pairs of mapping relationships between the associative words and their corresponding stroke information are constructed.

[0122] For the word data (i.e., associative words) with the search ranking of 20%-50%, only the stroke information of the first N characters in the associative words is constructed. If the number of characters in the associative word is less than or equal to N, the stroke information of the entire associative word is constructed.

[0123] For the last 50% of the word data (i.e. the associative words) after the search, only the first M characters in the associative words are constructed with stroke information. If the characters in the associative words are less than or equal to M, all stroke information is constructed for the associative words.

[0124] In this embodiment, N and M are both positive integers, and N>M.

[0125] It can be seen that by searching the data and differentially associating the stroke information of the words, it is avoided to generate all stroke-associated word pairs, which effectively saves resources and improves the efficiency of processing data.

[0126] As another embodiment of the present invention, before filtering the associative words at the top of the search data according to the associative word search data, the method further includes:

[0127] Obtain the log data of the search engine, and perform data cleaning on the log data to obtain valid log data and the reliability of word data.

[0128] Specifically, the cleaning rules are as follows:

[0129] 1. Remove the first 1% of data, which is crawler data;

[0130] 2. Remove data from blacklisted IDs;

[0131] 3. Remove data from blacklisted IPs;

[0132] 4. Remove data with abnormal operation behavior in a single day;

[0133] 5. Remove data whose source cannot be determined.

[0134] Figure 2 is a schematic diagram of the main process of a method for sorting associative words according to another reference embodiment of the present invention. The method for sorting associative words may include:

[0135] Step 201, obtaining log data, and performing data cleaning on the log data to obtain search data of associated words;

[0136] Step 202, based on the search data of the associative words, filter the associative words that are at the top of the search data;

[0137] Step 203, constructing stroke information corresponding to the associative word, and obtaining a mapping relationship between the associative word and each stroke information corresponding thereto;

[0138] Step 204, obtaining stroke information to be associated, and determining the heat value and business data value of each associated word corresponding to the stroke information;

[0139] Step 205, determining the stroke similarity between the stroke information and the strokes of each of the associated words;

[0140] Step 206: sort the associated words corresponding to the stroke information according to the heat value, the business data value and the stroke similarity.

[0141] It should be noted that step 204 may be performed first and then step 205, or step 205 may be performed first and then step 204, or step 204 and step 205 may be performed simultaneously, and the present invention does not limit this.

[0142] The method provided by the embodiment of the present invention solves the problem that the associative words corresponding to the stroke prefix stroke information to be associated cannot be sorted by adopting the technical means of sorting the associative words corresponding to the stroke prefix stroke information to be associated according to the heat value, the business data value and the stroke similarity. That is to say, the prior art generates all stroke-associative word pairs indiscriminately, resulting in a waste of resources. The present invention, however, comprehensively sorts the associative words corresponding to the stroke prefix stroke information to be associated according to the heat value, the business data value and the stroke similarity, thereby avoiding the indiscriminate generation of all stroke-associative word pairs. Therefore, the method provided by the embodiment of the present invention arranges each associative word in descending order according to the characteristics of the associative word itself (i.e., the heat value, the business data value and the stroke similarity), thereby generating stroke-associative word pairs differently, which can avoid the waste of resources and improve the speed of processing data.

[0143] In addition, the specific implementation content of the method for sorting associative words in a reference embodiment of the present invention has been described in detail in the method for sorting associative words described above, so the repeated content will not be described here.

[0144] Figure 3 is a schematic diagram of main modules of a device for sorting associative words according to an embodiment of the present invention, such as Figure 3 As shown, the device for sorting associative words includes a first determining module 301, a second determining module 302 and a sorting module 303. The first determining module 301 obtains stroke information to be associated, determines the heat value and business data value of each associative word corresponding to the stroke information, and the stroke information includes at least one stroke and / or at least one character; the second determining module 300 determines the stroke similarity between the stroke information and each associative word; the sorting module 303 sorts each associative word corresponding to the stroke prefix stroke information according to the heat value, business data and stroke similarity.

[0145] Specifically, the first determination module 301 obtains stroke information input by the user, that is, stroke information to be associated, determines each associated word corresponding to the stroke information to be associated, and finally determines the heat value and business data value of each associated word.

[0146] Optionally, the determining of the heat value of each associative word corresponding to the stroke information may include: calculating the Bayesian mean and the Newton cooling value respectively according to the number of searches for each associative word corresponding to the stroke information; calculating the heat value of each associative word corresponding to the stroke information according to the Bayesian mean and the Newton cooling value to ensure the validity and reliability of the heat value. Specifically, in this step, the Bayesian mean of each associative word may be first calculated by the Bayesian formula, and then the Newton cooling value of each associative word may be calculated by the Newton cooling formula, and finally the heat value of each associative word may be obtained based on the Bayesian mean and the Newton cooling value.

[0147] Optionally, the step of calculating the heat value may include:

[0148] First, the Bayesian average of each associative word is calculated by the Bayesian formula, and the Bayesian formula is B(wi)=(Atp(wi))+Tp(avg)*R(avg)) / (TpSum(wi)+Tp(avg)), i=0,1,2,…,m,

[0149] Where m is the total number of all associated words corresponding to the stroke information;

[0150] B(wi) is the Bayesian mean of the i-th associated word;

[0151] Atp(wi) is the current search count of the i-th associative word;

[0152] TpSum(wi) is the sum of the historical search times and the current search times of the i-th associative word;

[0153] Tp(avg) is the average of the total number of searches for all associated words;

[0154] R(avg) is the average value of Atp(wi) / TpSum(wi) of all associated words;

[0155] Then, the Newton cooling value of each associative word is calculated by the Newton cooling formula, and the Newton cooling formula is C(wi)=log(Atp(wi)+1 / Btp(wi)+1),

[0156] Among them, C(wi) is the Newton cooling value of the i-th associative word,

[0157] Btp(wi) is the number of historical searches for the i-th associative word;

[0158] Finally, the heat value of the i-th associative word is calculated according to the formula H(wi)=X*B(wi)+Y*C(wi), where H(wi) is the heat value of the i-th associative word, 0<X<1, 0<Y<1, and X+Y=1.

[0159] It should be noted that X and Y can be determined according to the weights of the Bayesian mean and the Newton cooling value, and different values ​​can be set as needed.

[0160] Therefore, the device provided by the embodiment of the present invention can accurately calculate the heat value of each associated word through the Bayesian formula and the Newton cooling formula, thereby providing effective and reliable basic data for subsequent steps. Moreover, using the heat value as the basis for sorting can more accurately sort the associated words corresponding to the stroke information, reducing the user's screening workload.

[0161] Optionally, determining the business data value of each associative word corresponding to the stroke information includes: calculating the business data value of each associative word according to the sales data, click-through rate and click-through conversion rate of each associative word and their respective weights. Specifically, the first determination module 301 obtains the sales data, click-through rate and click-through conversion rate corresponding to each associative word, and then calculates the business data value of each associative word based on these data and their respective weights.

[0162] Optionally, the business data value of each associative word is calculated based on different weight coefficients corresponding to sales data, click-through rate, and click-through conversion rate. For example, business data value = S*sales + P*click-through rate + Q*click-through conversion rate, where S is the weight coefficient of sales, P is the weight coefficient of click-through rate, and Q is the weight coefficient of click-through conversion rate. Accordingly, 0<S<1, 0<P<1, 0<Q<1, S+P+Q=1. S, P, and Q can be determined based on the weights of sales data, click-through rate, and click-through conversion rate, and different values ​​can be set as needed.

[0163] As another embodiment of the present invention, the sales volume is the normalized sales volume (nor_gmv), the click-through rate is the normalized click-through rate (nor_ctr), and the click-through conversion rate is the normalized click-through conversion rate (nor_cvr), so as to unify the results of these three indicators into the same scale, thereby improving the validity and reliability of the business data value.

[0164] Optionally, the second determination module 302 calculates the stroke similarity of each associated word according to the number of strokes included in the stroke information and the total number of strokes included in each associated word. The second determination module 302 obtains the number of strokes included in the stroke information and the total number of strokes included in each associated word, and then calculates the ratio of the number of strokes included in the stroke information to the total number of strokes included in a certain associated word. For example, stroke similarity=100*number of strokes included in the stroke information / total number of strokes included in the associated word. Therefore, the stroke similarity of the stroke information and the associated word is used as basic data to provide effective and reliable sorting for subsequent steps, and the stroke similarity can further improve the accuracy of the association between the stroke information and the associated word.

[0165] Optionally, the sorting module 303 calculates the total score of each associative word according to the weight of the heat value, business data value and stroke similarity, and sorts each associative word according to the total score of each associative word, so as to determine the priority of each associative word. Specifically, the total score of each associative word can be calculated by: total score = U*heat value + V*business data value + W*stroke similarity, wherein 0<U<100, 0<V<100, 0<W<100, U+V+W=100. U, V, and W can be determined according to the weight of heat value, business data value and stroke similarity, and different values ​​can be set as needed.

[0166] Optionally, the device further comprises a construction module, wherein the construction module selects the associative words with the highest search data according to the search data of the associative words, constructs the stroke information corresponding to the associative words, and obtains the mapping relationship between the associative words and the stroke information corresponding to the associative words. The construction module can determine the search ranking of the associative words according to the search data of the associative words, and thus construct the stroke information of the associative words differently according to the search ranking.

[0167] According to various embodiments described above, it can be seen that the present invention solves the problem that the associative words corresponding to the stroke information to be associated cannot be sorted by adopting the technical means of sorting the associative words corresponding to the stroke information to be associated according to the heat value, the business data value and the stroke similarity. In other words, the prior art generates all stroke-associative word pairs indiscriminately, resulting in a waste of resources. The present invention, however, comprehensively sorts the associative words corresponding to the stroke information to be associated according to the heat value, the business data value and the stroke similarity, thereby avoiding the indiscriminate generation of all stroke-associative word pairs. Therefore, the method provided in the embodiment of the present invention sorts the associative words according to the characteristics of the associative words themselves (i.e., heat value, business data value and stroke similarity), thereby generating stroke-associative word pairs differently, which can avoid the waste of resources and improve the speed of processing data.

[0168] It should be noted that the specific implementation content of the device for sorting associative words in the present invention has been described in detail in the method for sorting associative words described above, so the repeated content will not be described here.

[0169] The present invention also provides an information search method, which is applied to a server and includes:

[0170] receiving a search request from a user, wherein the search request carries stroke information to be associated;

[0171] Parsing the search request to obtain stroke information to be associated;

[0172] Determining a heat value and a business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character;

[0173] Determining the stroke similarity between the stroke information and the strokes of each of the associated words;

[0174] sorting the associated words corresponding to the stroke information according to the heat value, the business data value and the stroke similarity;

[0175] The sorting results of each associated word corresponding to the stroke information are pushed to the user.

[0176] Optionally, the associated words corresponding to the stroke information may be arranged in descending order, and then the first several associated words in the order may be pushed to the user, so that the multiple associated words with the highest correlation with the stroke information are displayed on the user interface, thereby facilitating user selection and improving user experience.

[0177] Optionally, determining the heat value of each associative word corresponding to the stroke information includes:

[0178] Calculating the Bayesian mean and the Newton cooling value respectively according to the number of searches for each associated word corresponding to the stroke information;

[0179] The heat value of each associative word corresponding to the stroke information is calculated according to the Bayesian mean value and the Newton cooling value.

[0180] Optionally, determining the heat value of each associative word corresponding to the stroke information includes:

[0181] The Bayesian average of each associative word is calculated by the Bayesian formula, where the Bayesian formula is B(wi)=(Atp(wi))+Tp(avg)*R(avg)) / (TpSum(wi)+Tp(avg)), i=0,1,2,…,m;

[0182] Wherein, m is the total number of all associated words corresponding to the stroke information, B(wi) is the Bayesian mean of the i-th associated word, Atp(wi) is the current search count of the i-th associated word, TpSum(wi) is the sum of the historical search count and the current search count of the i-th associated word, Tp(avg) is the average of the sum of the search counts of all associated words, and R(avg) is the average of Atp(wi) / TpSum(wi) of all associated words;

[0183] The Newton cooling value of each associative word is calculated by the Newton cooling formula, where the Newton cooling formula is C(wi)=log(Atp(wi)+1 / Btp(wi)+1);

[0184] Among them, C(wi) is the Newton cooling value of the i-th associative word, and Btp(wi) is the historical search times of the i-th associative word;

[0185] Calculate the heat value of the i-th associative word according to the formula H(wi)=X*B(wi)+Y*C(wi);

[0186] Among them, H(wi) is the heat value of the i-th associative word, 0<X<1, 0<Y<1, X+Y=1.

[0187] Optionally, determining the business data value of each associative word corresponding to the stroke information includes:

[0188] Normalizing the sales data, click-through rate and click-through conversion rate of each of the associated words respectively;

[0189] The business data value of each of the association words is calculated based on the normalized sales data, click-through rate and click-through conversion rate, and their respective weights.

[0190] Optionally, determining the stroke similarity between the stroke information and the strokes of each of the associated words includes:

[0191] The stroke similarity of each associated word is calculated according to the number of strokes included in the stroke information and the total number of strokes included in each associated word.

[0192] Optionally, before obtaining the stroke information to be associated and determining the heat value and business data value of each associated word corresponding to the stroke information, the method further includes:

[0193] According to the search data of the associated words, filter the associated words with the highest search data;

[0194] Stroke information corresponding to the associative word is constructed, and a mapping relationship between the associative word and each stroke information corresponding thereto is obtained.

[0195] It should be noted that the specific implementation content of the information search method of the present invention has been described in detail in the above-mentioned method for sorting associative words, so the repeated content will not be described again here.

[0196] Figure 4 An exemplary system architecture 400 is shown to which the method for sorting associative words or the apparatus for sorting associative words according to the embodiments of the present invention can be applied.

[0197] like Figure 4 As shown, system architecture 400 may include terminal devices 401, 402, 403, network 404 and server 405. Network 404 is used to provide a medium for communication links between terminal devices 401, 402, 403 and server 405. Network 404 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0198] Users can use terminal devices 401, 402, 403 to interact with server 405 through network 404 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 401, 402, 403, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0199] The terminal devices 401 , 402 , and 403 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0200] The server 405 may be a server that provides various services, such as a backend management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 401, 402, and 403. The backend management server may analyze and process the received data such as product information query requests, and feed back the processing results (such as target push information, product information - only an example) to the terminal device.

[0201] It should be noted that the method for sorting associative words provided in the embodiment of the present invention is generally executed on the terminal devices 401, 402, 403 in a public place, and can also be executed by the server 405. Accordingly, the device for sorting associative words is generally set on the terminal devices 401, 402, 403 in a public place, and can also be set in the server 405.

[0202] It should be understood that Figure 4 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0203] Reference below Figure 5 , which shows a schematic diagram of the structure of a computer system 500 of a terminal device suitable for implementing an embodiment of the present invention. Figure 5 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0204] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage part 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the system 500 are also stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0205] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed, so that a computer program read therefrom is installed into the storage section 508 as needed.

[0206] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 509, and / or installed from the removable medium 511. When the computer program is executed by the central processing unit (CPU) 501, the above-mentioned functions defined in the system of the present invention are executed.

[0207] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0208] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0209] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes a first determination module, a second determination module and a sorting module, wherein the names of these modules do not constitute a limitation on the modules themselves in some cases.

[0210] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: obtaining stroke information to be associated, determining the heat value and business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character; determining the stroke similarity between the stroke information and each associated word; and sorting each associated word corresponding to the stroke information according to the heat value, business data value and stroke similarity.

[0211] According to the technical solution of the embodiment of the present invention: because the technical means of sorting the associated words corresponding to the stroke information to be associated according to the heat value, the business data value and the stroke similarity is adopted, the technical problem of being unable to sort the associated words corresponding to the strokes is overcome, and the associated words corresponding to the stroke information to be associated are sorted comprehensively according to the heat value, the business data value and the stroke similarity, thereby avoiding the indiscriminate generation of all stroke-associative word pairs. Therefore, the method provided by the embodiment of the present invention sorts the associated words according to the characteristics of the associated words themselves, thereby generating stroke-associative word pairs differently, which can avoid waste of resources and improve the speed of processing data.

[0212] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for sorting associative words, characterized in that: include: Acquire stroke information to be associated, and determine the heat value and business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character; Determining the stroke similarity between the stroke information and the strokes of each of the associated words; sorting the associated words corresponding to the stroke information according to the heat value, the business data value and the stroke similarity; Determining the heat value of each associated word corresponding to the stroke information includes: Calculating the Bayesian mean and the Newton cooling value respectively according to the number of searches for each associated word corresponding to the stroke information; Calculating the heat value of each associative word corresponding to the stroke information according to the Bayesian mean and the Newton cooling value; Determining the business data value of each associative word corresponding to the stroke information includes: Normalizing the sales data, click-through rate and click-through conversion rate of each of the associated words respectively; The business data value of each of the association words is calculated based on the normalized sales data, click-through rate and click-through conversion rate, and their respective weights.

2. The method according to claim 1, characterized in that Determining the heat value of each associated word corresponding to the stroke information includes: The Bayesian average of each associative word is calculated by the Bayesian formula, and the Bayesian formula is: B(wi)=(Atp (wi))+Tp(avg)*R(avg)) / (TpSum(wi)+Tp(avg)) , i=0,1,2,…,m; Where m is the total number of all associated words corresponding to the stroke information. B(wi) For the i The Bayesian average of the associated words, Atp(wi) For the i The current search count of associated words, TpSum(wi) For the i The sum of the historical search times and the current search times of the associated words. Tp(avg) is the average of the total number of searches for all associated words. R(avg) For all associated words Atp(wi) / TpSum(wi) The average value of The Newton cooling value of each of the associated words is calculated by the Newton cooling formula, which is: C(wi)=log (Atp(wi)+1 / Btp(wi)+1) ; in, C(wi) For the i Newtonian cooling value of the associated words, Btp(wi) For the i The historical search times of the associated words; According to the formula H(wi)=X*B(wi)+Y*C(wi) Calculate the i The heat value of the associated words; in, H(wi) For the i The heat value of the associated word, 0< X <1,0< Y <1, X + Y= 1.

3. The method according to claim 1, characterized in that: Determining the stroke similarity between the stroke information and the strokes of each of the associated words includes: The stroke similarity of each associated word is calculated according to the number of strokes included in the stroke information and the total number of strokes included in each associated word.

4. The method according to claim 1, characterized in that: Before obtaining the stroke information to be associated and determining the heat value and business data value of each associated word corresponding to the stroke information, the method further includes: According to the search data of the associated words, filter the associated words with the highest search data; Stroke information corresponding to the associative word is constructed, and a mapping relationship between the associative word and each stroke information corresponding thereto is obtained.

5. A device for sorting associative words, characterized in that: include: A first determination module is used to obtain stroke information to be associated, and determine the heat value and business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character; A second determination module is used to determine the stroke similarity between the stroke information and the strokes of each of the associated words; A sorting module, used for sorting each associative word corresponding to the stroke information according to the heat value, the business data value and the stroke similarity; Determining the heat value of each associated word corresponding to the stroke information includes: Calculating the Bayesian mean and the Newton cooling value respectively according to the number of searches for each associated word corresponding to the stroke information; Calculating the heat value of each associative word corresponding to the stroke information according to the Bayesian mean and the Newton cooling value; Determining the business data value of each associative word corresponding to the stroke information includes: Normalizing the sales data, click-through rate and click-through conversion rate of each of the associated words respectively; The business data value of each of the association words is calculated based on the normalized sales data, click-through rate and click-through conversion rate, and their respective weights.

6. The device according to claim 5, characterized in that The second determining module is used to: The stroke similarity of each associated word is calculated according to the number of strokes included in the stroke information and the total number of strokes included in each associated word.

7. A method for information search, characterized in that: include: receiving a search request from a user, wherein the search request carries stroke information to be associated; Parsing the search request to obtain stroke information to be associated; Determining a heat value and a business data value of each associated word corresponding to the stroke information, wherein the stroke information includes at least one stroke and / or at least one character; Determining the stroke similarity between the stroke information and the strokes of each of the associated words; sorting the associated words corresponding to the stroke information according to the heat value, the business data value and the stroke similarity; Pushing the sorting results of each associated word corresponding to the stroke information to the user; Determining the heat value of each associated word corresponding to the stroke information includes: Calculating the Bayesian mean and the Newton cooling value respectively according to the number of searches for each associated word corresponding to the stroke information; Calculating the heat value of each associative word corresponding to the stroke information according to the Bayesian mean and the Newton cooling value; Determining the business data value of each associative word corresponding to the stroke information includes: Normalizing the sales data, click-through rate and click-through conversion rate of each of the associated words respectively; The business data value of each of the association words is calculated based on the normalized sales data, click-through rate and click-through conversion rate, and their respective weights.

8. The method according to claim 7, characterized in that Determining the heat value of each associated word corresponding to the stroke information includes: The Bayesian average of each associative word is calculated by the Bayesian formula, and the Bayesian formula is: B(wi)=(Atp (wi))+Tp(avg)*R(avg)) / (TpSum(wi)+Tp(avg)) , i=0,1,2,…,m; Where m is the total number of all associated words corresponding to the stroke information. B(wi) For the i The Bayesian average of the associated words, Atp(wi) For the i The current search count of associated words, TpSum(wi) For the i The sum of the historical search times and the current search times of the associated words. Tp(avg) is the average of the total number of searches for all associated words. R(avg) For all associated words Atp(wi) / TpSum(wi) The average value of The Newton cooling value of each of the associated words is calculated by the Newton cooling formula, which is: C(wi)=log (Atp(wi)+1 / Btp(wi)+1) ; in, C(wi) For the i Newtonian cooling value of the associated words, Btp(wi) For the i The historical search times of the associated words; According to the formula H(wi)=X*B(wi)+Y*C(wi) Calculate the i The heat value of the associated words; in, H(wi) For the i The heat value of the associated word, 0< X <1,0< Y <1, X + Y= 1.

9. The method according to claim 7, characterized in that: Determining the stroke similarity between the stroke information and the strokes of each of the associated words includes: The stroke similarity of each associated word is calculated according to the number of strokes included in the stroke information and the total number of strokes included in each associated word.

10. The method according to claim 7, characterized in that Before obtaining the stroke information to be associated and determining the heat value and business data value of each associated word corresponding to the stroke information, the method further includes: According to the search data of the associated words, filter the associated words with the highest search data; Stroke information corresponding to the associative word is constructed, and a mapping relationship between the associative word and each stroke information corresponding thereto is obtained.

11. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1-4 and 7-10.

12. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 and 7 to 10 is implemented.

Images