Icon

Google+GPT+Pubchem_​ChemSearch_​Public

Visual Analysis of Sales Data

Workflow that is built according to the instructions on the "Build a workflow" page and in the KNIME Quickstart Guide

Shows possible steps in analyzing (sales) data
1. Data access - Reading a csv file
2. Data preprocessing - Filtering columns and rows
3. Data visualization - Assigning colors to groups in the data. Building graphs.

Parse JSON response Make a request Use ChatGPT-3.5 to modify compound name.(or to do whatever) Issue Request to gpt-3.5-turbo Use pubchem REST API to retrieve SMILES based on GTP name search Use pubchem REST API to retrieve SMILES based on CAS from Google Use Google Search Custom API to obtain results with CAS number Use Google Search Custom API to obtain results with CAS number - use if thesearch makes errors Get compoundnamesEnd the loopExecute the requestGet the answerDecomposeStar loopto get values forall entriesJSON format for API Requestmax = 100 tokensGet JSONDouble-clickand copy+pasteyour API keyAssign the answer tothe original rowMake the requestAssign originalrowidDefine a ChatGPTrequestGet URL to make the searchCheck missingURLsFilter RetrievedColumsProcess JSONVisualize StructuresDefine SMILESName prepGet ResultsSplit noresultsRemove missingAdd originalIDGet URL to make a searchDefine smilesLoopEndRemove missingURLsExtract CASbased on"Title"Split incorrectCASIs CAScorrect?Make the searchGet ResultsID as variableExtract CASbased on"Summary"Remove processingcolumnsIs CAScorrect?Split incorrectCASFound structuresFilter redundantcolumnsMake a searchqueryGet values1 row per3 secondsWait for 3 secondsMerge dataGet the most commonresultGet entries w/o resultsRemoveprocessingcolumnsSplit toanamecnameAdd originalID1 row per3 secondsWait for 3 secondsID as variableMake a searchqueryMake the searchLoopEndCombineresultsWrite each resultcontinuouslyFilter onlystring columnsUse to continue froma broken connectionRead searchresultsExampledata Table Rowto Variable Loop End POST Request Column Filter Ungroup Group Loop Start ContainerInput (JSON) JSON Path Input API Key Joiner JSON Transformer ConstantValue Column String Manipulation(Variable) String Manipulation Missing Value Column Filter JSON to Table RDKit InteractiveTable Molecule Type Cast String Manipulation GET Request Row Splitter Row Filter ConstantValue Column String Manipulation Molecule Type Cast Loop End Missing Value String Manipulation Rule-basedRow Splitter String Manipulation Web Searcher GET Request Table Rowto Variable String Manipulation Column Filter String Manipulation Rule-basedRow Splitter RDKit InteractiveTable Column Filter String Manipulation JSON to Table Chunk Loop Start Wait... Concatenate GroupBy ReferenceRow Filter Column Filter Row Splitter ConstantValue Column Chunk Loop Start Wait... Table Rowto Variable String Manipulation Web Searcher Loop End Concatenate CSV Writer Column Filter Row Filter CSV Reader CSV Reader Parse JSON response Make a request Use ChatGPT-3.5 to modify compound name.(or to do whatever) Issue Request to gpt-3.5-turbo Use pubchem REST API to retrieve SMILES based on GTP name search Use pubchem REST API to retrieve SMILES based on CAS from Google Use Google Search Custom API to obtain results with CAS number Use Google Search Custom API to obtain results with CAS number - use if thesearch makes errors Get compoundnamesEnd the loopExecute the requestGet the answerDecomposeStar loopto get values forall entriesJSON format for API Requestmax = 100 tokensGet JSONDouble-clickand copy+pasteyour API keyAssign the answer tothe original rowMake the requestAssign originalrowidDefine a ChatGPTrequestGet URL to make the searchCheck missingURLsFilter RetrievedColumsProcess JSONVisualize StructuresDefine SMILESName prepGet ResultsSplit noresultsRemove missingAdd originalIDGet URL to make a searchDefine smilesLoopEndRemove missingURLsExtract CASbased on"Title"Split incorrectCASIs CAScorrect?Make the searchGet ResultsID as variableExtract CASbased on"Summary"Remove processingcolumnsIs CAScorrect?Split incorrectCASFound structuresFilter redundantcolumnsMake a searchqueryGet values1 row per3 secondsWait for 3 secondsMerge dataGet the most commonresultGet entries w/o resultsRemoveprocessingcolumnsSplit toanamecnameAdd originalID1 row per3 secondsWait for 3 secondsID as variableMake a searchqueryMake the searchLoopEndCombineresultsWrite each resultcontinuouslyFilter onlystring columnsUse to continue froma broken connectionRead searchresultsExampledata Table Rowto Variable Loop End POST Request Column Filter Ungroup Group Loop Start ContainerInput (JSON) JSON Path Input API Key Joiner JSON Transformer ConstantValue Column String Manipulation(Variable) String Manipulation Missing Value Column Filter JSON to Table RDKit InteractiveTable Molecule Type Cast String Manipulation GET Request Row Splitter Row Filter ConstantValue Column String Manipulation Molecule Type Cast Loop End Missing Value String Manipulation Rule-basedRow Splitter String Manipulation Web Searcher GET Request Table Rowto Variable String Manipulation Column Filter String Manipulation Rule-basedRow Splitter RDKit InteractiveTable Column Filter String Manipulation JSON to Table Chunk Loop Start Wait... Concatenate GroupBy ReferenceRow Filter Column Filter Row Splitter ConstantValue Column Chunk Loop Start Wait... Table Rowto Variable String Manipulation Web Searcher Loop End Concatenate CSV Writer Column Filter Row Filter CSV Reader CSV Reader

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: Maarit Laukkanen
Created at: 2018-07-23
On NodePit since: 2023-06-29
Last update: 2024-05-02
Created with KNIME version: v4.7.4
Tags: data accessrow filteringcolumn filteringdata visualizationquickstart guidebeginnergetting startedfirst workflow

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!