Genetic Algorithm-Based Matrix Factorization for Missing Value Prediction

2017 ◽  
pp. 504-516 ◽  
Author(s):  
Sujoy Chatterjee ◽  
Anirban Mukhopadhyay
Keyword(s):  
Genetic Algorithm ◽  
Matrix Factorization ◽  
Missing Value ◽  
Value Prediction ◽  
Missing Value Prediction

Enhance the Quality of Collaborative Filtering using Tagging

2020 ◽  
Vol 13 ◽  
Author(s):  
Latha Banda ◽  
Karan Singh
Keyword(s):  
Genetic Algorithm ◽  
Collaborative Filtering ◽  
Cold Start ◽  
Collaborative Tagging ◽  
Rating Data ◽  
Missing Value ◽  
Value Prediction ◽  
Sparsity Problem ◽  
Missing Value Prediction ◽  
Weight Model

Background: Due to enormous data in web sites, recommending users for every item is impossible. For this problem Recommender Systems (RS) are introduced. RS is categorized into content-based (CB), collaborative Filtering (CF) and Hybrid RS. Based on these techniques recommendations are done to user. In this, CF is the recent technique used in RS in which tagging features also provided. Objective: Three main issues occur in RS are scalability problem which occurs when there is a huge data, sparsity problem occurs when rating data is missing and cols start user or item problem occurs when new user or new item enters in the system. To avoid these issues here we have proposed Tag and Time weight model with GA in Collaborative Tagging. Method: Here we have proposed a method Collaborative Tagging (CT) with Tag and Time weight model with real value genetic algorithm which enhances the recommendation quality by removing the issues of sparsity and cold start user problems with the help of missing value prediction. Here in this the sparsity problem can be removed using missing value prediction and cold start problems are removed using tag and time weight model using GA. Results: Here we have compared the results of Collaborative Filtering with cosine similarity (CF-CS), Collaborative Filtering with Diffusion Similarity (CF-DS), Tag and Time weight model with Diffusion similarity (TAW-TIW-DS) and Tag and Time weight model using Diffusion similarity and Genetic algorithm (TAW-TIW-DS-GA). Conclusion: Here we have compare the proposed approach with the baseline approaches and the metrics are used MAE, prediction percentage, Hit-rate and Hit-rank. Based on these metrics for every split TAW-TIW-DS-GA shown best results as compared to existing approach.

scholarly journals Missing value prediction for qualitative information systems

Filomat ◽  
2020 ◽  
Vol 34 (1) ◽  
pp. 175-185
Author(s):  
T. Medhat ◽  
Manal Elsayed
Keyword(s):  
Information Systems ◽  
Qualitative Information ◽  
Missing Value ◽  
Value Prediction ◽  
Missing Value Prediction

Missing Value Imputation Using ANN Optimized by Genetic Algorithm

2018 ◽  
Vol 5 (2) ◽  
pp. 41-57 ◽  
Author(s):  
Anjana Mishra ◽  
Bighnaraj Naik ◽  
Suresh Kumar Srichandan
Keyword(s):  
Genetic Algorithm ◽  
Missing Data ◽  
Data Entry ◽  
Missing Information ◽  
Missing Value ◽  
Need Analysis ◽  
Processing Data ◽  
Real World Applications ◽  
Long Time ◽  
Almost All

Missing value arises in almost all serious statistical analyses and creates numerous problems in processing data in databases. In real world applications, information may be missing due to instrumental errors, optional fields and non-response to some questions in surveys, data entry errors, etc. Most of the data mining techniques need analysis of complete data without any missing information and this induces researchers to develop efficient methods to handle them. It is one of the most important areas where research is being carried out for a long time in various domains. The objective of this article is to handle missing data, using an evolutionary (genetic) algorithm including some relatively simple methodologies that can often yield reasonable results. The proposed method uses genetic algorithm and multi-layer perceptron (MLP) for accurately predicting missing data with higher accuracy.

Sign in / Sign up

Export Citation Format

Share Document