patphy Goto Github PK

bigdatapatent

Text Mining Patents for Big Data Course Project

claimanalysis

Messing around with automatically analysing patent claims

foxinyu.github.io

Patent Research & Analysis

lda

A Java package that does basic LDA, without hyperparameter optimization. Folder settings are local. Ymmv.

network-analysis-on-patent-data

pa-project

patent analysis and visualization project

patent-analysis-framework

Proposed framework for extracting and analyzing Patent data

patent-extractor

Proposed Innovation Strategy recommendations for Hewlett-Packard by performing data analysis on company’s patent data spanning over 15 years. Data analysis include evaluation of linear and logistic regression models after performing sentimental analysis on the data. Further, it was followed by topic modeling to get more insight on various Operational Divisions of Hewlett-Packard. These python programs are used to extract the patent data of HP from a Cluster into a single file for effective data analysis.

patent-retrieval-using-citation-analysis

patent_analysis_python

patentanalysis

Analysis of patents related to big data, artificial intelligence, machine learning, prediction, data analysis, etc.

patentanalysis-1

Several tools may be useful for patent analysis

patentanalysis-2

patent

pathunt

B2B Patent Analysis Tool

pubmed-patent-parser

Data parsing code for patent analysis project

relevance-ranking-using-latent-semantic-indexing-

Latent Semantic Analysis Introduction: An information retrieval technique patented in 1988. In the context of its application to information retrieval, it is sometimes called Latent Semantic Indexing (LSI). LSI allows a search engine to determine what a page is about outside of specifically matching search query text. It looks at “Themes” instead of “Keywords”. Linear Algebra techniques used in the project: Singular Value Decomposition, Cosine Similarity, Matrix properties. Dataset: “Sci.space” news group from 20 news groups dataset, available in the Scikit-Learn library. It contains 400 news articles related to space. SVD (Singular Value Decomposition): SVD is a matrix decomposition algorithm, it decomposes a matrix into 3 matrices which are a set to transformations. Decomposition leads to an orthogonal matrix U, Diagonal matrix S and a Diagonal Matrix V. This is the best possible transformation of a matrix. In this decomposition method we are looking for a set of orthonormal basis in the row space that when multiplied by the original matrix goes to an orthonormal basis in the column space.Av1 = σ1u1  Av2 = σ2u2 

trend-analysis-based-patent-evaluation

My thesis project

ucl_msc_thesis_2016

Repo for all code related to my 2016 dissertation in semantic patent analysis at UCL.