No self-introduction
Loading
Profile
Follow
codes (6)
Recommendation system - Collaborative Filtering Al
no vote
Application background Java implementation of several recommendation algorithm: SVD RSVD, ItemNeighborSVD, slopeone because just to see the effect of RMSE, so many functions are not perfectThis code is what I see from the Internet, I feel very useful for my learning recommendation algorithm, so put on everyone to share, I hope to be useful to everyone. Key Technology If you need to run to see the effect then you need to1 download the movielens data set, the proposed use of data set http://www.grouplens.org/sites/www.grouplens.org/external_files/data/ml-10m.zip; 1000W2 view modify regeneration.py to generate a data set again. Then modify the format of the DataConverter generation algorithm based on common3 according to the need to modify the code in the test directory, run. Parameter configuration can be the first do not move,
py_qijing 0 0
1
In action mahout book source code implementation
no vote
Application background Recently in learning mahout, bought a Book mahout in action. From books online download the source code implementation practice enhance my impression, very good code, very useful. Pass up to share with you, hope to help everyone. Key Technology After the decompression inside ReadmeInstall thrift, mavenPom.xml <: exec;Executable= "/usr/local/bin/thrift" > < ExecExecutable= "/usr/local/bin/thrift/bin/thrift" > and the thrift installation position inside the quotes..Pom.xml thrift version from 0.6.1 to 0.9.0, because classifier.java is generated by 0.9.0 version.Eclipse:eclipse MVNAt this time, the project has a red exclamation, add environmental variables: M2_REPO= your.M2/repository path can.Import will project eclipse in.Continue to operate according to the method of readme.
py_qijing 1 0
1
Recommended system code
4.0
Application background Recently in the learning recommendation system, from the site to see this code, download and run very useful, for me to learn the recommendation system, understand the recommendation system is a great help,The recommendation algorithm used in the movie recommendation system is based on the collaborative filtering algorithm (CollaborativeRecommendation Filtering). Key Technology Download the Data on the GroupLens website (http://www.grouplens.org/node/12).Sets, in the film system, we have used nearly 900 users for 1683 of the movie review the data set of nearly 100000 rows. Will download the ml-data_0.zip score data and movie information data out. Converting the score file to a text file similar to the CSV file format, CSV is comma separated value file (CommaValue Separated) is a plain text file format used to store data, file named rating.TXT.
py_qijing 0 0
1
Movie recommendation system source code
4.0
Application background File contains installation notes and some instructions for those ThisWant to get started quickly. Much more detailed documentation on whoTo setup and program CoFE is available in the Docs subdirectory. how Key Technology Requirements: System1 or above j2sdk1.4.12 database 3.* or above MySQLCoFE comes distributed with the MySQL client libraries Note:2.X in lib/3rdParty/mysql-connectorJ-2.0.14/mysql.jar. If you want to use the 3.X ConnectorJ ConnectorJYou, can download them yourself from www.mysql.com libraries, butWill need to update the ant build files youPossibly bin/start_server.sh to point to the new library. and
py_qijing 0 0
1
Recommendation system - Collaborative Filtering Al
no vote
Application background This program is based on the Java implementation of the user based collaborative filtering algorithm, the algorithm is simple, due to the online information on the implementation of the code rarely,So the main purpose of the information is to provide a reference for your information, for reference only. Key Technology U1.base and u1.test are training sets and test sets, respectively, from MovieLens data sets,This procedure is simply based on the user's collaborative filtering algorithmThe configuration information required by the running algorithm, including the read training set and the test set, and the nearest number of the nearest neighbor number, can be found in the Base.java file.The main program of this program is Application.javaFor reference only, I hope to help
py_qijing 3 0
1
Movie recommendation system source code
5.0
Application background This is the project code of this blog, MyEclipse engineering file. After download, you can directly import MyEclipse to run.Since the original project was developed in the mahout reference jar MyEclipse package, so this compression file and no relevant jar file,So run this project before you need to do, is the need to import mahout development kits. Key Technology Mahout Apache is Software Foundation Apache (ASF)An open source program, to provide some of the classic algorithm to learn the field of machine learning, designed to help developers more convenient and quickly create intelligent applications. The classical algorithm includes clustering, classification and cooperation.Filter, evolutionary programming, and so on, and in the Mahout also added support for Hadoop Apache, so that these algorithms can run more efficient in the cloud computing environment
py_qijing 3 0
1
No more~
Follows0
Fans0