CN103631779A - Word recommending system based on socialized dictionary - Google Patents
Word recommending system based on socialized dictionary Download PDFInfo
- Publication number
- CN103631779A CN103631779A CN201210298147.4A CN201210298147A CN103631779A CN 103631779 A CN103631779 A CN 103631779A CN 201210298147 A CN201210298147 A CN 201210298147A CN 103631779 A CN103631779 A CN 103631779A
- Authority
- CN
- China
- Prior art keywords
- word
- user
- dictionary
- recommended
- socialized
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a word recommending system based on a socialized dictionary. While a user looks up words in a network dictionary, the words are provided for the socialized dictionary to generate a word bank which is to be recommended and used for selection through a new word extraction algorithm; the word bank to be recommended forms a recommending word bank through a filtering integrating method; according to user behaviors, collaborative filtering is performed, personalized recommending word banks are generated, classified in a personalizing mode according to the word bank to be recommend and recommended to the user. The word recommending system has the advantages that according to the scheme, the personalized word learning method is adopted, so that pointed and efficient word learning is provide for the user effectively and rapidly. In addition, a social network can be effectively used for finding pointed words, so that learning quality is improved.
Description
Technical field
The present invention relates to mobile interconnected field, particularly the word commending system of social activityization dictionary.
Background technology
User, use in the process of dictionary, learning word, how according to each user's self feature, recommending word targetedly, optimize its structure of knowledge, improve the specific aim of learning, the efficiency that improves study, is individualized learning and the major issue that individualized education is provided.And existing personalized word study, just according to fixing dictionary, the study after a kind of basic classification that the user of different levels type is carried out.This sorting technique is too simple, is not easy to the knowledge that user's continuous learning is new, and vocabulary is not done to effective classification yet, and therefore personalized recommendation word has larger room for improvement.
In addition, because existing personalized word sorting technique is not considered the learning and communication of socialization, the study that user is isolated, may cause following problems: one like this: the study that user can not be in good time is to new word; Its two: can not effectively share the knowledge that arrives of study, its three: learning efficiency is low.
By retrieval, find, the patent No. is CN2010102007365, and name is called a kind of method and application terminal thereof of personalized meta search.This patent is a kind of method of personalized search, comprising: set up in advance interest storehouse; According to definite searched keyword, from interest storehouse, extract usage log, and the Search Results that search engine is returned carries out pre-service; Utilize usage log, the interest-degree of the URLs (URL) of corresponding described searched keyword in pretreated Search Results and interest storehouse is calculated, according to result of calculation, sort and show; Upgrade described interest storehouse.
The patent No. is a kind of ordering system and the method that CN2006100363617 name is called integrative searching result, this patent is a kind of ordering system of integrative searching result, for the result of the separate vertical search engine of comprehensive search engine is sorted, described system comprises data analysis module, database and order module; Described data analysis module provides the data message that can be used for sort algorithm, and imports in described database and store; Described database is used for storing the data message that data analysis module provides, the extraction while supplying described order module execution sort algorithm, and store the final sequencing information that described order module obtains; Described order module is carried out sort algorithm for the data message of storing according to described database, described vertical search engine is sorted, and final sequencing information is stored in described database.
The patent No. is that CN2007100995946 name is called a kind of search engine retrieving result reordering method based on user behavior information, this patent is a kind of search engine retrieving result reordering method based on user behavior information, the method is to realize according to the following steps successively on the computing machine of search engine: step (1). the screening of the conventional query set of user: step (1.1). and data pre-service: extract from the user journal of at least one search engine by search engine web server for carrying out based on user behavior the user profile that Search Results reorders, the user profile forming at least comprises following content: Query: the inquiry that user submits to, URL: this inquires about the results page address that corresponding user clicks, Id: identification number when system is automatically distributed to each user and used search engine at every turn, described step (1.1) contains following sub-step successively: step (1.1.1). the coded format of search engine web server record is converted to the GBK form of Chinese characters of the national standard coding, step (1.1.2). remove described Query, URL, Id information in addition, and log information is organized into the form of three content item character strings of described user journal, step (1.1.3). in the scope of step (1.1.2), recycling, as the noise information in the string matching algorithm filter user inquiry of KMP, only retains the directly content item of reflection search engine common user query demand behavior, step (1.2). extract inquiring user and count information: each the inquiry Q submitting to for the user in user journal in the nearest time period setting, statistics was submitted the number of users of this inquiry Q to, this numeric representation the attention rate of user to this inquiry, step (1.3). the screening of conventional query set: if: certain inquiry Q its inquiring user number in search engine user daily record is less than setting value, gets rid of outside conventional query set, otherwise, this inquiry Q is placed in described conventional query set S, step (2). the extraction of user's clicking rate information: step (2.1). the extraction of single search engine user clicking rate: user's clicking rate=for inquiry Q, user clicks user's the click total degree of number of times/inquiry Q of results page address URL, step (2.2). under multiple search engine, user's clicking rate information merges, with a probability expression P (URL| inquires about Q), be illustrated in the user's clicking rate to the results page address URL of user's clicking rate of inquiry Q after merging: P (URL| inquire about Q)=* P(SE ↓ [ i ] | inquiry Q). P (URL|SE ↓ [ i ] inquires about Q), wherein, P (SE ↓ [ i ] | inquiry Q) is illustrated in the probability of inquiring about Q in i search engine SE ↓ [ i ], by SE ↓ [ i ] inquiry confidence level, represent: SE ↓ [ i ] inquiry confidence level=log(inquire about the total number of users of Q among search engine logs SE ↓ [ i ]) total number of users of the middle Q of inquiry of/* log(SE ↓ [ i ]), i=1,2 ..., IP (URL|SE ↓ [ i ], inquiry Q) be illustrated in search engine logs SE ↓ [ i ], for inquiry Q.
Above patent is all that Search Results is resequenced, and be user-friendly to, but existing patent just lays particular emphasis on the sequence of Webpage searching result.
Summary of the invention
For the deficiencies in the prior art, the present invention solves and pushes which word to user, thereby is convenient to the problem of user learning, provides a kind of solution of personalization, by means of the use of social activity dictionary, for different users recommends word targetedly.The object of the present invention is to provide and a kind ofly can effectively utilize social networks education resource, provide personalized word to recommend learning method.
To achieve these goals, the technical solution used in the present invention is:
A word commending system based on social activity dictionary, user is when Web-Based Dictionary is looked into word, and the word providing to social activity dictionary, by new word extraction algorithm, generates a selective word library to be recommended;
Word library to be recommended forms recommendation word library by filtering integrated approach;
According to user behavior, carry out collaborative filtering, produce personalized recommendation word library, for word dictionary to be recommended, carry out personalization classification, be pushed to user.
Social activityization dictionary: user in using the process of dictionary, being submitted to social networks and exchanging of automatic or manual.
Extract word to be recommended: user, use in the reciprocal process of socialization dictionary, collect the new word of user, comprise each word and relative all words of user of each user relatively.
User's classification: according to the service condition of user's log-on message, social activityization dictionary, by clustering algorithm, user is classified.
Filter integrated word to be recommended: word to be recommended is carried out to denoising, the operation such as integrated.
Recommend dictionary: what certain stage formed can be for the dictionary of recommending.
Recommend word: on social activity dictionary interface, show the word of recommending.
This patent lays particular emphasis on the sequence to dictionary, and the sequence to a plurality of lexical or textual analysis in dictionary
Beneficial effect of the present invention: when the present invention adopts such scheme that personalized word learning method is provided, can be fast and effectively to user provide targetedly, word study efficiently; In addition, the present invention can effectively utilize social networks, finds that there is word targetedly, thereby improves the quality of study.
Accompanying drawing explanation
Fig. 1, functional-block diagram of the present invention.
Embodiment
Whole functional-block diagram as of the present invention in Fig. 1; Registered user uses social activityization dictionary after logging in, and used word in recording user social activityization dictionary extracts neologisms from used word, is saved in word library to be recommended; Word library to be recommended is carried out to the processing such as denoising, generating recommendations word library; According to user's behavior, carry out collaborative filtering generation for each user's personalized recommendation word library; In suitable, recommend user.
Claims (1)
1. the word commending system based on social activity dictionary, is characterized in that,
User is when Web-Based Dictionary is looked into word, and the word providing to social activity dictionary, by new word extraction algorithm, generates a selective word library to be recommended;
Word library to be recommended forms recommendation word library by filtering integrated approach;
According to user behavior, carry out collaborative filtering, produce personalized recommendation word library, for word dictionary to be recommended, carry out personalization classification, be pushed to user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210298147.4A CN103631779A (en) | 2012-08-21 | 2012-08-21 | Word recommending system based on socialized dictionary |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210298147.4A CN103631779A (en) | 2012-08-21 | 2012-08-21 | Word recommending system based on socialized dictionary |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103631779A true CN103631779A (en) | 2014-03-12 |
Family
ID=50212855
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210298147.4A Pending CN103631779A (en) | 2012-08-21 | 2012-08-21 | Word recommending system based on socialized dictionary |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103631779A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942317A (en) * | 2014-04-25 | 2014-07-23 | 江西迈思科技有限公司 | Recommending method and system |
CN108108346A (en) * | 2016-11-25 | 2018-06-01 | 广东亿迅科技有限公司 | The theme feature word abstracting method and device of document |
US11710054B2 (en) * | 2012-10-08 | 2023-07-25 | Tencent Technology (Shenzhen) Company Limited | Information recommendation method, apparatus, and server based on user data in an online forum |
-
2012
- 2012-08-21 CN CN201210298147.4A patent/CN103631779A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11710054B2 (en) * | 2012-10-08 | 2023-07-25 | Tencent Technology (Shenzhen) Company Limited | Information recommendation method, apparatus, and server based on user data in an online forum |
CN103942317A (en) * | 2014-04-25 | 2014-07-23 | 江西迈思科技有限公司 | Recommending method and system |
CN103942317B (en) * | 2014-04-25 | 2017-05-03 | 江西迈思科技有限公司 | Recommending method and system |
CN108108346A (en) * | 2016-11-25 | 2018-06-01 | 广东亿迅科技有限公司 | The theme feature word abstracting method and device of document |
CN108108346B (en) * | 2016-11-25 | 2021-12-24 | 广东亿迅科技有限公司 | Method and device for extracting theme characteristic words of document |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11580104B2 (en) | Method, apparatus, device, and storage medium for intention recommendation | |
CN103914478B (en) | Webpage training method and system, webpage Forecasting Methodology and system | |
CN103365924B (en) | A kind of method of internet information search, device and terminal | |
CN106156127B (en) | Method and device for selecting data content to push to terminal | |
CN102096717B (en) | Search method and search engine | |
CN101593200B (en) | Method for classifying Chinese webpages based on keyword frequency analysis | |
CN101416179B (en) | System and method for providing regulated recommended word to every subscriber | |
CN100405371C (en) | Method and system for abstracting new word | |
CN102184204B (en) | Auto fill method and system of intelligent Web form | |
CN102117317B (en) | Blind person Internet system based on voice technology | |
CN105005600B (en) | Preprocessing method of URL (Uniform Resource Locator) in access log | |
CN100394727C (en) | Log analyzing method and system | |
CN103023714B (en) | The liveness of topic Network Based and cluster topology analytical system and method | |
CN102819591B (en) | A kind of content-based Web page classification method and system | |
CN103544255A (en) | Text semantic relativity based network public opinion information analysis method | |
CN101685521A (en) | Method for showing advertisements in webpage and system | |
CN102053983A (en) | Method, system and device for querying vertical search | |
WO2014180130A1 (en) | Method and system for recommending contents | |
CN103389998A (en) | Novel Internet commercial intelligence information semantic analysis technology based on cloud service | |
CN105677787B (en) | Information retrieval device and information search method | |
CN104123366A (en) | Search method and server | |
CN104077407A (en) | System and method for intelligent data searching | |
CN110188291B (en) | Document processing based on proxy log | |
CN102722499A (en) | Search engine and implementation method thereof | |
CN102737021A (en) | Search engine and realization method thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140312 |