CN103198079B - The implementation method of relevant search and device - Google Patents
The implementation method of relevant search and device Download PDFInfo
- Publication number
- CN103198079B CN103198079B CN201210004943.2A CN201210004943A CN103198079B CN 103198079 B CN103198079 B CN 103198079B CN 201210004943 A CN201210004943 A CN 201210004943A CN 103198079 B CN103198079 B CN 103198079B
- Authority
- CN
- China
- Prior art keywords
- search
- relevant
- character string
- word
- check
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000011218 segmentation Effects 0.000 claims description 9
- 229910052739 hydrogen Inorganic materials 0.000 claims description 6
- 230000008878 coupling Effects 0.000 claims description 3
- 238000010168 coupling process Methods 0.000 claims description 3
- 238000005859 coupling reaction Methods 0.000 claims description 3
- 230000013011 mating Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 241000234295 Musa Species 0.000 description 4
- 235000018290 Musa x paradisiaca Nutrition 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a kind of implementation method of relevant search, comprising: the search need character string obtaining user's input; Search need character string participle is obtained multiple queries word; The relevant in order search listing of each query word is obtained in inverted index; The relevant in order search listing of each query word is returned to user.The invention provides a kind of implement device of relevant search, comprising: acquisition module, for obtaining the search need character string of user's input; Word-dividing mode, for obtaining multiple queries word by search need character string participle; List block, for obtaining the relevant in order search listing of each query word in inverted index; Submit module to, for the relevant in order search listing of each query word is returned to user.The present invention can provide relevant search to user expeditiously.
Description
Technical field
The present invention relates to search technique field, in particular to a kind of implementation method and device of relevant search.
Background technology
Along with the rapid expanding of internet information, the degree of dependence of user to search engine is increasing, and its commercial value cannot be estimated.Therefore, the status of search engine correlation technique is important all the more, particularly has material impact to the analysis and understanding of user search demand to Search Results.Generally speaking, cause search effect not good, the query word often due to user's selection is incorrect, if system can provide relevant search need for the search need of user, user therefrom will obtain some and inspire, and the Product Experience of user will be improved significantly.
Relevant search, that some are relevant with current search demand, but the relevant search need provided by other user or dictionary, in realization, be generally according to the searched popular degree of query word and and user-selected query word between similarity, by what produce after system automatic decision.Relevant search is often arranged in left side or the below of search results pages.On the internet hunt websites such as Google, relevant search is applied, but its implementation is not open.
Summary of the invention
The present invention aims to provide a kind of implementation method and device of relevant search, to provide relevant search to user.
In an embodiment of the present invention, provide a kind of implementation method of relevant search, comprising: the search need character string obtaining user's input; Search need character string participle is obtained multiple queries word; The relevant in order search listing of each query word is obtained in inverted index; The relevant in order search listing of each query word is returned to user;
The relevant in order search listing obtaining query word described in each in inverted index comprises: mate described inverted index with current described query word; The described search need character string of each record described is obtained as described relevant search list from the described address that the described property value matched is corresponding; To search need character string marking Score=α * T+ β * H+ γ * D+ δ * M described in each in described relevant search list ordering; Wherein, the α weight that to be weight, the β arranged T be is arranged H, γ are the weight, the δ that arrange D is the weight arranged M, inputs number of times, D is described search need character string inputs number of times in one day, M is described search need character string inputs number of times in one month in one hour that T is total input number of times of described search need character string, H is described search need character string.
In an embodiment of the present invention, provide a kind of implement device of relevant search, comprising: acquisition module, for obtaining the search need character string of user's input; Word-dividing mode, for obtaining multiple queries word by search need character string participle; List block, for obtaining the relevant in order search listing of each query word in inverted index; Submit module to, for the relevant in order search listing of each query word is returned to user;
Described list block, for mating described inverted index with current described query word; The described search need character string of each record described is obtained as described relevant search list from the described address that the described property value matched is corresponding; To search need character string marking Score=α * T+ β * H+ γ * D+ δ * M described in each in described relevant search list ordering; Wherein, the α weight that to be weight, the β arranged T be is arranged H, γ are the weight, the δ that arrange D is the weight arranged M, inputs number of times, D is described search need character string inputs number of times in one day, M is described search need character string inputs number of times in one month in one hour that T is total input number of times of described search need character string, H is described search need character string.
The implementation method of the relevant search of the above embodiment of the present invention and device adopt inverted structure, can provide relevant search expeditiously to user.
Accompanying drawing explanation
Accompanying drawing described herein is used to provide a further understanding of the present invention, and form a application's part, schematic description and description of the present invention, for explaining the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 shows the process flow diagram of the implementation method of the relevant search according to the embodiment of the present invention;
Fig. 2 shows the schematic diagram set according to the Trie of the embodiment of the present invention;
Fig. 3 shows the schematic diagram of the inverted index according to the embodiment of the present invention;
Fig. 4 shows the schematic diagram of the implement device of the relevant search according to the embodiment of the present invention.
Embodiment
Below with reference to the accompanying drawings and in conjunction with the embodiments, describe the present invention in detail.
Fig. 1 shows the process flow diagram of the implementation method of the relevant search according to the embodiment of the present invention, comprising:
Step S10, obtains the search need character string of user's input;
Step S20, obtains multiple queries word by search need character string participle;
Step S30, obtains the relevant in order search listing of each query word in inverted index;
Step S40, returns to user by the relevant in order search listing of each query word.
This method adopts inverted structure, can provide relevant search expeditiously to user.
Preferably, step S20 comprises: build dictionary for word segmentation in advance; Adopt two-way maximum matching method by search need string matching dictionary for word segmentation; Disambiguation dictionary is utilized to carry out word sense disambiguation the inconsistent part of coupling.The present embodiment realizes easily via computer programming.Utilize Chinese words segmentation, search need is changed into the query word that several are concrete, be convenient to carry out subsequent treatment.
Preferably, adopt two array algroithm to build dictionary for word segmentation, specifically comprise:
1) create Trie tree according to the entry of resource dictionary, such as, the dictionary creation module creation Trie that on market, the upright intelligence of current public offering is thought in intelligent analysis system 4.1 editions can be used to set;
2) create even numbers group, comprising:
Integer array base [] and check [] is formed
3) in being set by Trie, the child nodes of root node adds queue, wherein, sorts from big to small to the child nodes number that the element in queue has according to it;
4) first element of queue is taken out;
5) if this element has child nodes, then to the character value B1 of all child nodes of this element, B2 ... Bn, get one and meet check [H+B1], check [H+B2] ... check [H+Bn] is the value H of 0, once find this H, base [i]=H is then set, check [H+B1]=check [H+B2]=...=check [H+Bn]=i, wherein, H+B1, H+B2 ... H+Bn is the subscript position of current child nodes in even numbers group, i is the subscript position of the father node of current child nodes, if the ending character of this element representation entry, base [i]=-H is then set,
6) if this element does not have child nodes, then base [i]=-i is set;
7) step 4-6 is repeated, until all elements all takes in queue;
8) dictionary for word segmentation is formed with even numbers group.
Adopt the dictionary of the storage mode of even numbers group can determine an entry fast whether in dictionary, also can know its subscript position.Such as, entry " 2345 ", retrieving is as follows: base [0]+2=2, check whether base [2] is negative, be not negative, then base [2]+3=2+3=5, checks whether base [5] is negative, is not negative, then base [5]+4=5+4=9, check whether base [9] is negative, is not negative, then base [9]+5=9+5=14, check whether base [14] is negative, be found to be negative and equal-14, illustrating and have found word " 2345 ", and be designated as 14 under knowing the data at the end of this word.The value of base [2], base [3], base [4], base [5] is assume that in this example.
Trie tree is also known as dictionary tree, word lookup tree, and be a kind of tree structure, for preserving a large amount of character strings, Fig. 2 shows the schematic diagram set according to the Trie of the embodiment of the present invention.Its fundamental property is:
1, root node does not comprise character, and except root node, each node only comprises a character.
2, from root node to a certain node, on path, the Connection operator of process gets up, and is the character string that this node is corresponding.
3, the character that comprises of all child nodes of each node is not identical.
The present embodiment realizes easily via computer programming.
Preferably, this method also comprises establishment inverted index, and it comprises multiple item, and every comprises the address that a property value has each record of property value, and each property value records a query word respectively, and each record is respectively the search need character string obtained for each time.Fig. 3 shows the schematic diagram of the inverted index according to the embodiment of the present invention.The present embodiment realizes easily via computer programming.
Inverted index (English: Invertedindex), also be often called as reverse indexing, insert archives or reverse archives, be a kind of indexing means, be used to be stored in the mapping of the memory location of certain word in a document or one group of document under full-text search.It is data structure the most frequently used in DRS.There is the inverted index form that two kinds are different:
Article one, the horizontal inverted index (or arranging file index) of record comprises the list of the document of each reference words.
The horizontal inverted index (or complete inverted index) of a word comprises again the position of each word in a document.The form of the latter provides more compatibility (such as phrase search), but needs more Time and place to create.
Example is for English, and here wants indexed text:
T0＝"itiswhatitis"
T1＝"whatisit"
T2＝"itisabanana"
We just can obtain inverted file index below:
"a":{2}
"banana":{2}
"is":{0,1,2}
"it":{0,1,2}
"what":{0,1}
The condition " what " of retrieval, " is " and " it " is by this set of correspondence:.
To identical word, obtain these complete inverted indexs below, the paired data be made up of the word result of number of documents and current queries.Equally, the word result of number of documents and current queries is all started from scratch.So, " banana ": { (2,3) } in other words " banana ", in the 3rd document (T2), and are the 4th word (address are 3) in the position of the 3rd document.
"a":{(2,2)}
"banana":{(2,3)}
"is":{(0,1),(0,4),(1,1),(2,1)}
"it":{(0,0),(0,3),(1,2),(2,0)}
"what":{(0,2),(1,0)}
If execution phrase search " whatisit ", the whole words result place document separately obtaining this phrase is document 0 and document 1.But the continuous print condition of this phrase retrieval only obtains at document 1.
Application inverted index data structure is typical search engine retrieving algorithm part and parcel.The target of a search engine execution is exactly the speed of Optimizing Queries: find the place that certain word occurs in a document.In the past, forward index developed the list of the word for storing each document, then turned around to develop a kind of inverted index.The inquiry of forward index often meets the orderly full-text query frequently of each document and each word is verifying the such inquiry of the checking in document.
In fact, the restriction of time, internal memory, processor etc. resource, technical forward index is irrealizable.In order to the word list of each document of alternative forward index, the inverted index data structure listing the list of word all places document of each inquiry develops out.Along with the establishment of inverted index, inquiry of today indicates by word immediately and obtains rapidly result (through storing at random).Random storage is also considered to usually faster than sequential storage.
Preferably, step S30 comprises: with current query word coupling inverted index; The search need character string of each record is obtained as relevant search list from the address that the property value matched is corresponding; According to the attribute of each search need character string to relevant search list ordering.The present embodiment realizes easily via computer programming.
Preferably, according to the attribute of each search need character string, relevant search list ordering is comprised: to each search need character string marking Score=α * T+ β * H+ γ * D+ δ * M in relevant search list ordering; Wherein, the α weight that to be weight, the β arranged T be is arranged H, γ are the weight, the δ that arrange D is the weight arranged M, inputs number of times, D is search need character string inputs number of times in one day, M is search need character string inputs number of times in one month in one hour that T is total input number of times of search need character string, H is search need character string.The present embodiment realizes easily via computer programming.
Preferably, this method also comprises with even numbers group and attribute array structure relevant search feedback dictionary, obtains T, H, D and M, wherein, create even numbers group and specifically comprise from relevant search feedback dictionary:
1) Trie tree is created according to each search need character string;
2) create initial even numbers group, comprising:
Integer array base [] and check [] is formed;
3) in being set by Trie, the child nodes of root node adds queue, wherein, sorts from big to small to the child nodes number that the element in queue has according to it;
4) first element of queue is taken out;
5) if this element has child nodes, then to the character value B1 of all child nodes of this element, B2 ... Bn, get one and meet check [H+B1], check [H+B2] ... check [H+Bn] is the value H of 0, once find this H, base [i]=H is then set, check [H+B1]=check [H+B2]=...=check [H+Bn]=i, wherein, H+B1, H+B2 ... H+Bn is the subscript position of current child nodes in even numbers group, i is the subscript position of the father node of current child nodes, if the ending character of this element representation entry, base [i]=-H is then set,
6) if this element does not have child nodes, then base [i]=-i is set;
7) step 4-6 is repeated, until all elements all takes in queue;
Create attribute array to comprise:
1) history word frequency attribute T, H, D and M of statistical query word;
2) create initial attribute array, length equals the length of even numbers group;
3) all search need character strings are traveled through, return its array index value i in even numbers group;
4) be point to the pointer of T, H, D and M to attribute array [i] assignment.
The present embodiment realizes easily via computer programming.The present embodiment reconstructs while dictionary utilizing even numbers group, and add a parallel attribute array, parallel is say that the length of this array increased is the same with even numbers group.While utilizing even numbers group to judge whether this entry, according to subscript position, in the parallel array increased, obtain relative attribute.
Preferably, step S40 comprises: the relevant in order search listing of each query word is merged into a relevant search list; Heapsort is carried out in the relevant search list be combined; User is submitted in relevant search list after heapsort.
In an embodiment of the present invention, as follows to the heapsort method flow of the relevant in order search listing of each query word:
1) there will be a known several arrays, each array is made up of the relevant in order search listing of each query word.
2) operating value getting the current mark bit of each array builds a heap.Time initial, current mark bit points to first element of array, and the heap built is unordered.
3) utilize Heap algorithm to sort to this heap, get heap top operating value and put into results set.And move after the current mark bit of the array at operating value place, heap top.
4) second step is repeated, until all array current mark bit point to last element of this array.
The present embodiment realizes easily via computer programming.
Preferably, user is submitted in the relevant search list after heapsort to comprise: duplicate removal process is carried out in the relevant search list after heapsort; Submit the top n search need character string in the relevant search list after duplicate removal process to user, N is default integer.Such as obtained in relevant search, including AB search need by query word A, and in the relevant search obtained by query word B, also include AB search need, then can there is the search need of repetition in lists, removing the relevant search demand repeated can make result more accurate, to meet the individual demand of user.
Fig. 4 shows the schematic diagram of the implement device of the relevant search according to the embodiment of the present invention, comprising:
Acquisition module 10, for obtaining the search need character string of user's input;
Word-dividing mode 20, for obtaining multiple queries word by search need character string participle;
List block 30, for obtaining the relevant in order search listing of each query word in inverted index;
Submit module 40 to, for the relevant in order search listing of each query word is returned to user.
This device adopts inverted structure, can provide relevant search expeditiously to user.
Obviously, those skilled in the art should be understood that, above-mentioned of the present invention each module or each step can realize with general calculation element, they can concentrate on single calculation element, or be distributed on network that multiple calculation element forms, alternatively, they can realize with the executable program code of calculation element, thus, they can be stored and be performed by calculation element in the storage device, or they are made into each integrated circuit modules respectively, or the multiple module in them or step are made into single integrated circuit module to realize.Like this, the present invention is not restricted to any specific hardware and software combination.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (7)
1. an implementation method for relevant search, is characterized in that, comprising:
Obtain the search need character string of user's input;
Described search need character string participle is obtained multiple queries word;
The relevant in order search listing of query word described in each is obtained in inverted index, wherein, described inverted index comprises multiple item, every address of each record comprising a property value and there is described property value, property value described in each records a described query word respectively, and each record is respectively the described search need character string obtained for each time;
The relevant in order search listing of query word described in each is returned to user;
The relevant in order search listing obtaining query word described in each in inverted index comprises:
Described inverted index is mated with current described query word;
The described search need character string of each record described is obtained as described relevant search list from the described address that the described property value matched is corresponding;
To search need character string marking Score=α * T+ β * H+ γ * D+ δ * M described in each in described relevant search list ordering;
Wherein, the α weight that to be weight, the β arranged T be is arranged H, γ are the weight, the δ that arrange D is the weight arranged M, inputs number of times, D is described search need character string inputs number of times in one day, M is described search need character string inputs number of times in one month in one hour that T is total input number of times of described search need character string, H is described search need character string.
2. method according to claim 1, is characterized in that, described search need character string participle is obtained multiple queries word and comprises:
Build dictionary for word segmentation in advance;
Adopt two-way maximum matching method by dictionary for word segmentation described in described search need string matching;
Disambiguation dictionary is utilized to carry out word sense disambiguation the inconsistent part of coupling.
3. method according to claim 2, is characterized in that, adopts two array algroithm to build dictionary for word segmentation, specifically comprises:
1) Trie tree is created according to the entry of resource dictionary;
2) create even numbers group, comprise integer array base [] and check [] and form, the element in check [] and base [] array is initially set to 0 value, and array index initial value is 0;
3) child nodes of root node in described Trie tree is added queue, wherein, the child nodes number that the element in described queue has according to it is sorted from big to small;
4) first element of queue is taken out;
5) if this element has child nodes, then to the character value B1 of all child nodes of this element, B2 ... Bn, get one and meet check [H+B1], check [H+B2] ... check [H+Bn] be 0 on the occasion of H, once find this H, base [i]=H is then set, check [H+B1]=check [H+B2]=...=check [H+Bn]=i, wherein, H+B1, H+B2 ... H+Bn is the subscript position of current child nodes in even numbers group, i is the subscript position of the father node of current child nodes, if the ending character of this element representation entry, base [i]=-H is then set,
6) if this element does not have child nodes, then base [i]=-i is set;
7) step 4 is repeated)-6), until all elements all takes in described queue;
8) described dictionary for word segmentation is formed with described even numbers group.
4. method according to claim 1, is characterized in that, also comprises and builds relevant search feedback dictionary with even numbers group and attribute array, obtains T, H, D and M, wherein, create described even numbers group and specifically comprise from described relevant search feedback dictionary:
1) Trie tree is created according to search need character string described in each;
2) create initial even numbers group, comprising: integer array base [] and check [] is formed;
3) child nodes of root node in described Trie tree is added queue, wherein, the child nodes number that the element in described queue has according to it is sorted from big to small;
4) first element of queue is taken out;
5) if this element has child nodes, then to the character value B1 of all child nodes of this element, B2 ... Bn, get one and meet check [H+B1], check [H+B2] ... check [H+Bn] is the value H of 0, once find this H, base [i]=H is then set, check [H+B1]=check [H+B2]=...=check [H+Bn]=i, wherein, H+B1, H+B2 ... H+Bn is the subscript position of current child nodes in even numbers group, i is the subscript position of the father node of current child nodes, if the ending character of this element representation entry, base [i]=-H is then set,
6) if this element does not have child nodes, then base [i]=-i is set;
7) step 4 is repeated)-6), until all elements all takes in described queue; Create described attribute array to comprise:
1) history word frequency attribute T, H, D and M of described query word is added up;
2) create initial attribute array, length equals the length of described even numbers group;
3) all search need character strings are traveled through, return its array index value i in described even numbers group;
4) be point to the pointer of T, H, D and M to described attribute array [i] assignment.
5. method according to claim 1, is characterized in that, the relevant in order search listing of query word described in each is returned to user and comprises:
The relevant in order search listing of query word described in each is merged into a relevant search list;
Heapsort is carried out to the relevant search list of described merging;
User is submitted in relevant search list after described heapsort.
6. method according to claim 5, is characterized in that, user is submitted in the relevant search list after described heapsort and comprises:
Duplicate removal process is carried out in relevant search list after described heapsort;
Submit search need character string described in the top n in the relevant search list after described duplicate removal process to user, N is default integer.
7. an implement device for relevant search, is characterized in that, comprising:
Acquisition module, for obtaining the search need character string of user's input;
Word-dividing mode, for obtaining multiple queries word by described search need character string participle;
List block, for obtaining the relevant in order search listing of query word described in each in inverted index, wherein, described inverted index comprises multiple item, every address of each record comprising a property value and there is described property value, property value described in each records a described query word respectively, and each record is respectively the described search need character string obtained for each time;
Submit module to, for the relevant in order search listing of query word described in each is returned to user;
Described list block, for mating described inverted index with current described query word; The described search need character string of each record is obtained as described relevant search list from the address that the property value matched is corresponding; To search need character string marking Score=α * T+ β * H+ γ * D+ δ * M described in each in described relevant search list ordering;
Wherein, the α weight that to be weight, the β arranged T be is arranged H, γ are the weight, the δ that arrange D is the weight arranged M, inputs number of times, D is described search need character string inputs number of times in one day, M is described search need character string inputs number of times in one month in one hour that T is total input number of times of described search need character string, H is described search need character string.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210004943.2A CN103198079B (en) | 2012-01-06 | 2012-01-06 | The implementation method of relevant search and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210004943.2A CN103198079B (en) | 2012-01-06 | 2012-01-06 | The implementation method of relevant search and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103198079A CN103198079A (en) | 2013-07-10 |
| CN103198079B true CN103198079B (en) | 2016-04-20 |
Family
ID=48720648
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210004943.2A Expired - Fee Related CN103198079B (en) | 2012-01-06 | 2012-01-06 | The implementation method of relevant search and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103198079B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021202B (en) * | 2014-06-16 | 2017-11-24 | 百度在线网络技术（北京）有限公司 | The entry processing unit and method of a kind of knowledge sharing platform |
| CN104166550A (en) * | 2014-08-13 | 2014-11-26 | 扬州大学 | Software maintenance oriented method for re-customizing modification request |
| CN104598583B (en) * | 2015-01-14 | 2018-01-09 | 百度在线网络技术（北京）有限公司 | The generation method and device of query statement recommendation list |
| CN105224624A (en) * | 2015-09-22 | 2016-01-06 | 广州神马移动信息科技有限公司 | A kind of method and apparatus realizing down the quick merger of row chain |
| CN105912696A (en) * | 2016-04-25 | 2016-08-31 | 中国互联网络信息中心 | DNS (Domain Name System) index creating method and query method based on logarithm merging |
| CN107784027A (en) * | 2016-08-31 | 2018-03-09 | 北京国双科技有限公司 | A kind of reminding method and device of judgement document's search key |
| CN108021569A (en) * | 2016-11-01 | 2018-05-11 | 中国移动通信有限公司研究院 | The structure of AC automatic machines and Chinese multi-model matching method and relevant apparatus |
| CN108388480B (en) * | 2017-02-03 | 2021-06-11 | 百度在线网络技术（北京）有限公司 | Short string correlation verification method and device |
| CN108874880B (en) * | 2018-05-04 | 2021-11-23 | 昆明理工大学 | Trie-based space keyword query method and device |
| CN108846013B (en) * | 2018-05-04 | 2021-11-23 | 昆明理工大学 | Space keyword query method and device based on geohash and Patricia Trie |
| CN111104583B (en) * | 2018-10-10 | 2024-01-05 | 河南星易网络科技有限公司 | Live broadcast room recommendation method, storage medium, electronic equipment and system |
| CN109800412A (en) * | 2018-12-10 | 2019-05-24 | 鲁东大学 | A kind of Chinese word segmentation and big data information retrieval method and device |
| CN111008625B (en) * | 2019-12-06 | 2023-07-18 | 建信金融科技有限责任公司 | Address correction method, device, equipment and storage medium |
| CN112100132B (en) * | 2020-09-24 | 2024-07-30 | 深圳软牛科技有限公司 | Deleted file type identification method and device, electronic equipment and storage medium |
| CN113127662A (en) * | 2021-04-13 | 2021-07-16 | 广联达科技股份有限公司 | Component searching method and device, electronic equipment and readable storage medium |
| CN113326267B (en) * | 2021-06-24 | 2023-08-08 | 长三角信息智能创新研究院 | Address matching method based on inverted index and neural network algorithm |
| CN116303405B (en) * | 2023-05-12 | 2023-11-10 | 深圳竹云科技股份有限公司 | Data duplicate checking method and device and computer equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101192237A (en) * | 2006-11-30 | 2008-06-04 | 国际商业机器公司 | Method and system for inquiring multiple information |
| CN102023989A (en) * | 2009-09-23 | 2011-04-20 | 阿里巴巴集团控股有限公司 | Information retrieval method and system thereof |
| CN102110123A (en) * | 2009-12-29 | 2011-06-29 | 中国人民解放军国防科学技术大学 | Method for establishing inverted index |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7149748B1 (en) * | 2003-05-06 | 2006-12-12 | Sap Ag | Expanded inverted index |
-
2012
- 2012-01-06 CN CN201210004943.2A patent/CN103198079B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101192237A (en) * | 2006-11-30 | 2008-06-04 | 国际商业机器公司 | Method and system for inquiring multiple information |
| CN102023989A (en) * | 2009-09-23 | 2011-04-20 | 阿里巴巴集团控股有限公司 | Information retrieval method and system thereof |
| CN102110123A (en) * | 2009-12-29 | 2011-06-29 | 中国人民解放军国防科学技术大学 | Method for establishing inverted index |
Non-Patent Citations (1)
| Title |
|---|
| 网络教学汉语答疑系统关键技术的研究与应用;王红雨;《中国优秀硕士学位论文全文数据库 信息科技辑》;20060415(第4期);第20页第2.6.1节至第22页第2.6.3节,第33-35页第3.2.5节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103198079A (en) | 2013-07-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103198079B (en) | The implementation method of relevant search and device | |
| Bhagavatula et al. | Methods for exploring and mining tables on wikipedia | |
| CN1728142B (en) | Phrase identification method and device in an information retrieval system | |
| CN102084363B (en) | A method for efficiently supporting interactive, fuzzy search on structured data | |
| US9342582B2 (en) | Selection of atoms for search engine retrieval | |
| CN110851722A (en) | Search processing method, device and equipment based on dictionary tree and storage medium | |
| JP5746426B2 (en) | Discovery of index documents | |
| US8296279B1 (en) | Identifying results through substring searching | |
| US8620907B2 (en) | Matching funnel for large document index | |
| US20070244863A1 (en) | Systems and methods for performing searches within vertical domains | |
| US9805035B2 (en) | Systems and methods for multimedia image clustering | |
| CN106503223B (en) | online house source searching method and device combining position and keyword information | |
| US9569525B2 (en) | Techniques for entity-level technology recommendation | |
| NO338518B1 (en) | Multiple register-based information retrieval system | |
| US20070244862A1 (en) | Systems and methods for ranking vertical domains | |
| CN104715064A (en) | Method and server for marking keywords on webpage | |
| Wang et al. | Exploring traversal strategy for web forum crawling | |
| Hon et al. | Efficient index for retrieving top-k most frequent documents | |
| US20150347590A1 (en) | System and method for performing a pattern matching search | |
| Kempe | Structure and dynamics of information in networks | |
| CN105426490A (en) | Tree structure based indexing method | |
| Endres et al. | Index structures for preference database queries | |
| Radelaar et al. | Improving search and exploration in tag spaces using automated tag clustering | |
| Kantorski et al. | Choosing values for text fields in web forms | |
| Mor et al. | Effective presentation of results using ranking & clustering in meta search engine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220620 Address after: 3007, Hengqin international financial center building, No. 58, Huajin street, Hengqin new area, Zhuhai, Guangdong 519031 Patentee after: New founder holdings development Co.,Ltd. Patentee after: Peking University Patentee after: BEIJING FOUNDER ELECTRONICS Co.,Ltd. Address before: 100871, Beijing, Haidian District Cheng Fu Road 298, founder building, 5 floor Patentee before: PEKING UNIVERSITY FOUNDER GROUP Co.,Ltd. Patentee before: Peking University Patentee before: BEIJING FOUNDER ELECTRONICS Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160420 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |