CN108021658B - Intelligent big data searching method and system based on whale optimization algorithm - Google Patents

Abstract

The invention discloses a big data intelligent search method and system based on a whale optimization algorithm. The whale optimization algorithm is used to optimize and solve the big data intelligent search problem, thereby quickly obtaining the search results closest to the user's needs, and further improving the performance of the intelligent search engine. Accuracy and efficiency; the present invention can find a high-quality feasible solution to the big data intelligent search problem within an acceptable time cost. According to the user's search conditions, the big data related to the keyword input by the user is obtained from the engine database, and the user Continuously score the search results, and gradually obtain the search effect that meets the individual needs of users. The present invention uses the whale optimization algorithm to search big data intelligently, establishes an intelligent search engine based on ontology features, and quickly and efficiently finds the search results that best meet the user's needs, solving the problem that the current search engine cannot provide users with a good sense of The problem of interest information and high search efficiency.

Description

A Big Data Intelligent Search Method and System Based on Whale Optimization Algorithm

technical field

The invention belongs to the cross-application field of big data and intelligent computing, and relates to the application of an intelligent optimization algorithm in the field of big data, in particular to a solution to the problem of intelligent search of big data, and in particular to a whale-based optimization algorithm Big data intelligent search method and system.

Background technique

In recent years, big data has developed rapidly around the world, and at the same time it has attracted great attention from academics, industry and governments. By effectively managing big data and obtaining its value through analysis, high value-added applications and services can be provided to achieve more economic and social value. However, the advent of the era of big data has brought both major development opportunities and technical challenges. Traditional computing technology will face many technical difficulties when solving big data problems. Therefore, it is necessary to research and find new effective technical methods to complete the analysis and processing of big data and value discovery.

Different from traditional database queries, one of the characteristics of web search is that users often cannot fully express their needs at once, but only through multiple interactions with search engines and gradually approaching, can they basically achieve the goals requested by users. This can lead to very low efficiency. The search system can gain insight into the meaning of the query expression through big data analysis when the user makes the query request for the first time, and automatically extend the semantics of the query expression, which can greatly improve efficiency and reduce the burden on users in one go.

The web search engine is a powerful tool for online information query and a key technology of web information retrieval. Traditional search engines mainly include: catalog search engines, full-text search engines, meta search engines, collection search engines and vertical search engines. With the diversification of information formats and the surge in the amount of information, traditional search engines are facing enormous challenges, and they can no longer meet the more personalized, intelligent and diverse needs of users. From information collection to information organization and indexing, information retrieval and user interface, intelligent search engines are constantly optimizing all aspects of traditional search engines. Intelligent search engines mainly include: ontology-based intelligent search engines, knowledge base system-based intelligent search engines and semantic association-based intelligent search engines. An intelligent search engine is a search engine that improves the traditional search engine based on keyword level retrieval to knowledge or concept level retrieval. It can understand words from the perspective of knowledge and concepts, showing strong intelligence and personalization characteristics , it has a high natural language understanding and knowledge processing ability based on a certain knowledge base technology.

After more than 20 years of development, the big data search engine has greatly improved in search speed and accuracy, but its basic structure and technology have not changed much, which has caused its limitations. Search engines for different applications require different information. Non-intelligent retrieval lacks the ability to identify information that users are interested in, and the sorting method cannot be adjusted according to different users. Providing targeted search services for different people is one of the roads for the development of search engines in the future, so various intelligent search technologies were born. The birth of disciplines such as artificial intelligence and computational intelligence is that researchers try to get inspiration from human thinking and some laws of the biological world, create corresponding computing models, and apply them to information science. Artificial neural networks, swarm intelligence, and genetic computing are some of the successful examples applied to current big data challenges. Combining ontology features to build an intelligent search engine not only solves the problem that current search engines cannot provide users with interesting information, but also improves search efficiency.

Contents of the invention

In order to solve the problem of big data intelligent search, the present invention proposes a big data intelligent search method and system based on whale optimization algorithm.

The technical solution adopted by the method of the present invention is: a large data intelligent search method based on a whale optimization algorithm, characterized in that it comprises the following steps:

n表示维度，N表示鲸群大小；Step 1: Read in the user's search conditions, and get the big data matching the keywords entered by the user from the engine database according to the user's search conditions. Each big data is a whale, and the current position of the i-th whale in the whale group is X _i , initialize the position of the whale group:

n represents the dimension, and N represents the size of the whale group;

Step 2: Initialize the parameters required by the whale optimization algorithm, including the size of the whale group N, the logarithmic spiral shape constant b, the current iteration number j, the maximum number of iterations M, and the global optimal position of the entire whale group is G;

Step 3: Calculate the fitness function value of the initial position of the whale group in the whale optimization algorithm, and use the big data with the highest fitness function value as the best spatial position of the current whale group individual

Step 4: Calculate coefficient vectors A and C;

Step 5: Generate a random number p with a value range of [0,1], and choose different ways to update the spatial position of the whale group according to the value of p;

Step 6: Decode the updated position vector of the whale group into corresponding big data and present it to the user. The user scores the obtained big data according to their own search conditions as the fitness function value; find and save the best whale in the current group group individual X ^* ;

Step 7: Determine the position of the next generation of whales by comparing the fitness function values corresponding to the position vectors of the whales before and after the update;

Step 8: Record the position of the whale group corresponding to the big data with the highest degree of conformity as the global optimal solution G and its fitness function value;

Step 9: Determine whether the user has found the desired text document in the engine;

If not, then make j=j+1 and go back to step 4;

If so, then output the optimal whale group individual fitness value and the big data corresponding to the position X ^* .

The technical solution adopted by the system of the present invention is: it is characterized in that: comprising input module, whale optimization algorithm initialization module, fitness function value module, coefficient vector calculation module, whale group spatial position update mode selection module, updated whale group Spatial position vector fitness value calculation module, position determination module of the next generation whale group, global optimal solution G for the position of the whale group, and its fitness function value recording module and judgment module;

n表示维度，N表示鲸群大小；Described input module: be used for reading in user's search condition, obtain the big data that matches with user's input keyword from engine database according to user's search condition, each big data is a whale, the ith whale in the whale group The current position X _i , initialize the position of the whale group:

n represents the dimension, and N represents the size of the whale group;

Described whale optimization algorithm initialization module: be used for initializing the required parameter of whale optimization algorithm, comprise whale group size N, logarithmic spiral shape constant b, current number of iterations j, maximum number of iterations M, the global optimum position of whole group of whales is G;

The fitness function value module: used to calculate the fitness function value of the initial position of the whale group in the whale optimization algorithm, and use the big data with the highest evaluation of the fitness function value as the best spatial position of the current whale group individual

The coefficient vector calculation module: used to calculate coefficient vectors A and C;

The whale group spatial position update mode selection module: used to generate a random number p whose value range is [0,1], and select different ways to update the whale group spatial position according to the value of p;

The updated whale group spatial position vector fitness value calculation module: used to decode the updated position vector of the whale group into corresponding big data to present to the user, and the user scores the obtained big data according to his own search conditions, As the fitness function value; find and save the best whale group individual X ^* in the current group;

The position determination module of the next-generation whale group: for determining the position of the next-generation whale group by comparing the fitness function value corresponding to the position vector of the whale group before and after updating;

The position of the group of whales is the global optimal solution G and its fitness function value recording module: the position of the group of whales corresponding to the big data with the highest degree of conformity is the global optimal solution G and its fitness function value;

The judging module: used to judge whether the user has found the desired text document in the engine;

If not, then make j=j+1 and go back to step 4;

If so, then output the optimal whale group individual fitness value and the big data corresponding to the position X ^* .

The beneficial effects of the present invention are: the whale optimization algorithm has the characteristics of few adjustment parameters, fast convergence speed, high optimization precision, strong global optimization ability and good convergence stability, and the whale optimization algorithm is used to optimize and solve big data intelligent search problems, Therefore, the search results closest to the user's needs can be quickly obtained, with good accuracy and efficiency.

Description of drawings

Fig. 1 is a flow chart of the method of the embodiment of the present invention.

Detailed ways

In order to facilitate those of ordinary skill in the art to understand and implement the present invention, the present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the implementation examples described here are only used to illustrate and explain the present invention, and are not intended to limit this invention.

Please see Fig. 1, a kind of big data intelligent search method based on whale optimization algorithm provided by the present invention, comprises the following steps:

n表示维度，N表示鲸群大小；Step 1: Read in the user's search conditions, and get the big data matching the keywords entered by the user from the engine database according to the user's search conditions. Each big data is a whale, and the current position of the i-th whale in the whale group is X _i , initialize the position of the whale group:

n represents the dimension, and N represents the size of the whale group;

Step 2: Set the parameters required for the whale optimization algorithm. The parameters required for the initialization of the whale optimization algorithm include the size of the whale group N, the logarithmic spiral shape constant b, the current number of iterations j, the maximum number of iterations M, and the global value of the entire whale group The best position is G.

Step 3: Decode the initial position vector of the whale group in the whale optimization algorithm into corresponding big data and present it to the user. The user scores the obtained big data according to their own search conditions, which is the fitness function value. The engine takes the big data with the highest evaluation of the fitness function value as the best spatial position of the current whale population

Step 4: Calculate coefficient vectors A and C.

The calculation formula of the coefficient vector A is:

A=2a×r-a

Among them, M is the maximum number of iterations, and r is a random vector with a value range between [0,1].

The calculation formula of the coefficient vector C is:

C=2r

Among them, r is a random vector with a value range of [0,1].

Step 5: Generate a random number p with a value range of [0,1], and choose different ways to update the spatial position of the whale group according to the value of p.

When p<0.5, if A<1, the formula for updating the spatial position of the current individual whale group is:

X _j+1 ＝X _j -A×D

为当前鲸群个体最佳空间位置。Among them, j is the current number of iterations, X _j is the spatial position of the current individual whale group, A and C are coefficient vectors,

is the best spatial position of the current whale group individual.

When p<0.5, if A≥1, randomly select the position X _rand of the whale group individual from the current group, and update the spatial position of the current whale group individual. The formula for updating the spatial position of the current individual whale group is:

X＝X _rand -A×D

D＝|C×X _rand,j -X _j |

Among them, X _rand is a randomly selected position in the current whale group, that is, a random whale group individual; X _rand,j is a randomly selected position in the jth dimension of the current whale group;

When p≥0.5, the formula for updating the spatial position of the current whale population is:

为当前鲸群个体最佳空间位置。Among them, D′ is the distance between the current best position of the i-th whale in the whale group and the prey, b is the defined logarithmic spiral shape constant, l is a random number between [-1,1], X _j is the current The individual spatial position of the whale group,

is the best spatial position of the current whale group individual.

Step 6: Decode the updated position vector of the whale group into corresponding big data and present it to the user. The user scores the obtained big data according to their own search conditions as the fitness function value. Find and save the best whale group individual X ^* in the current group.

Step 7: Determine the position of the next generation of whales by comparing the fitness function values corresponding to the position vectors of the whales before and after the update. The determination rule is: if the fitness function value corresponding to the position vector of the updated whale group is higher than that before the update, replace the original whale group; otherwise, keep the whale group before the update. The calculation method of the fitness function value is the same as step 3.

Step 8: Record the position of the whale group corresponding to the big data with the highest degree of conformity as the global optimal solution G and its fitness function value.

Step 9: Determine whether the user has found the desired text document in the engine;

If not, then make j=j+1 and go back to step 4;

If so, then output the optimal whale group individual fitness value and the big data corresponding to the position X ^* .

The invention uses the whale optimization algorithm to optimize and solve the big data intelligent search problem, thereby quickly obtaining the search result closest to the user's needs, and the method can be used in the technical fields related to big data and intelligent computing.

It should be understood that the parts not described in detail in this specification belong to the prior art.

It should be understood that the above-mentioned descriptions for the preferred embodiments are relatively detailed, and should not therefore be considered as limiting the scope of the patent protection of the present invention. Within the scope of protection, replacements or modifications can also be made, all of which fall within the protection scope of the present invention, and the scope of protection of the present invention should be based on the appended claims.

Claims (7)

1. The big data intelligent searching method based on whale optimization algorithm is characterized by comprising the following steps:

step 1: reading in the search condition of the user, obtaining big data matched with the keyword input by the user from an engine database according to the search condition of the user, wherein each big data is one whale, and the current position X of the ith whale in the whale group _i Initializing the position of whale groups:

i=1, 2,..n, N represents the dimension, N represents the whale size;

step 2: parameters required by initializing a whale optimization algorithm comprise whale group size N, logarithmic spiral shape constant b, current iteration times j and maximum iteration times M, and the global optimal position of the whole whale group is G;

step 3: calculating the fitness function value of the initial position of the whale group in the whale optimization algorithm, and taking the big data with the highest fitness function value evaluation as the optimal spatial position of the current whale group individual

Step 4: calculating coefficient vectors A and C;

step 5: generating a random number p with a value range of [0,1], and selecting different ways for updating the space position of whale group according to the value of p;

step 6: decoding the position vector of the updated whale group into corresponding big data, and displaying the corresponding big data to a user, wherein the user scores the obtained big data according to the search condition of the user, and the big data is used as an fitness function value; finding and preserving the best whale individuals X in the current population ^* ；

Step 7: determining the position of the next generation whale group by comparing fitness function values corresponding to the position vectors of the whale groups before and after updating;

step 8: recording the whale group position corresponding to the big data with highest coincidence degree as a global optimal solution G and an fitness function value thereof;

step 9: judging whether the user finds a required text document in the engine;

if not, j=j+1 is caused and the step 4 is executed in a revolving way;

if yes, outputting the optimal whale individual fitness value and the position X thereof ^* Corresponding big data.

2. The whale optimization algorithm-based big data intelligent search method according to claim 1, wherein the method comprises the following steps: in step 3, the fitness function value is calculated by decoding the position vector of the whale group into corresponding big data and displaying the corresponding big data to the user, and the user scores the obtained big data according to the search condition.

3. The intelligent big data searching method based on whale optimization algorithm according to claim 1, wherein the calculation formula of the coefficient vector a in step 4 is:

A＝2a×r-a

wherein r is a random vector with a value range of 0 and 1.

4. The intelligent big data searching method based on whale optimization algorithm according to claim 1, wherein the calculation formula of the coefficient vector C in step 4 is:

C＝2r

wherein r is a random vector, and the value range is [0,1].

5. The intelligent big data searching method based on whale optimization algorithm according to claim 1, wherein the method of updating whale space position in step 5 is as follows:

when p is less than 0.5, if A is less than 1, updating the formula of the spatial position of the current whale individual to be:

X _j+1 ＝X _j -A×D

wherein j is the current iteration number, X _j For the current whale individual spatial position, a and C are coefficient vectors,

the optimal spatial position for the current whale individual;

when p is less than 0.5, if A is more than or equal to 1, randomly selecting whale group individual position X from the current group _rand Updating the spatial position of the current whale group individual; updating the spatial position formula of the current whale individual as follows:

X＝X _rand -A×D

D＝|C×X _rand,j -X _j |

wherein X is _rand Randomly selected positions in the current whale, namely random whale individuals; x is X _rand,j Randomly selecting a position for the j-th dimension in the current whale;

when p is more than or equal to 0.5, updating the spatial position formula of the current whale group individual as follows:

wherein D' is the distance from the current best position of the ith whale of the whale group to the prey, b is a defined logarithmic spiral shape constant, and l is a random number between [ -1, 1].

6. The intelligent big data searching method based on whale optimization algorithm according to any one of claims 1-5, wherein the rule for determining the position of next generation whale group in step 7 is: if the fitness function value corresponding to the position vector of the whale group after updating is higher than that before updating, the original whale group is replaced; otherwise, preserving whale groups before updating; the method for calculating the fitness function value is the same as the step 3.

7. A big data intelligent search system based on a whale optimization algorithm is characterized in that: the system comprises an input module, a whale optimization algorithm initialization module, an fitness function value module, a coefficient vector calculation module, a whale space position updating mode selection module, an updated whale space position vector fitness value calculation module, a next generation whale position determination module, a whale position being a global optimal solution G, a fitness function value recording module and a judgment module;

the input module: for reading in the search condition of the user, obtaining big data matched with the key words input by the user from the engine database according to the search condition of the user, wherein each big data is one whale, and the current position X of the ith whale in the whale group _i Initializing the position of whale groups:

i=1, 2,..n, N represents the dimension, N represents the whale size;

the whale optimization algorithm initialization module: parameters required for initializing a whale optimization algorithm comprise whale group size N, logarithmic spiral shape constant b, current iteration times j and maximum iteration times M, and the global optimal position of the whole whale group is G;

the fitness function value module: the method comprises the steps of calculating fitness function values of initial positions of whales in a whale optimization algorithm, and taking big data with highest fitness function value evaluation as the optimal spatial position of a current whale individual

The coefficient vector calculation module: for calculating coefficient vectors a and C;

the whale space position updating mode selection module: a mode for generating a random number p with a value range of [0,1] and selecting different updated whale group space positions according to the value of p;

the updated whale group space position vector fitness value calculating module is used for: the method comprises the steps of decoding the position vector of the whale group after updating into corresponding big data and showing the corresponding big data to a user, and scoring the obtained big data as a fitness function value by the user according to the searching condition of the user; finding and preserving the best whale individuals X in the current population ^* ；

The position determining module of the next generation whale group: the method comprises the steps of determining the position of a next generation whale group by comparing fitness function values corresponding to position vectors of whale groups before and after updating;

the whale group position is the global optimal solution G and the fitness function value recording module thereof: the whale group position corresponding to the big data with highest coincidence degree is recorded as a global optimal solution G and an fitness function value thereof;

the judging module is used for: for determining whether the user has found the desired text document in the engine;

if not, j=j+1 is caused and the step 4 is executed in a revolving way;

if yes, outputting the optimal whale individual fitness value and the position X thereof ^* Corresponding big data.

Images