🍪 Would you care for a cookie?

We use our own and third-party cookies to analyze our services according to profiling based on your browsing habits. You can learn more and set up your preferences here.

PreferencesAccept Cookies
Accept CookiesPreferences

🍪 Would you care for a cookie?

We use our own and third-party cookies to analyze our services according to profiling based on your browsing habits. You can learn more and set up your preferences here.

PreferencesAccept Cookies
Accept CookiesPreferences

Text Analysis

Text Analysis

Analyzing large quantities of text is a window into public opinion on a topic.

Deploy powerful natural language processing to discover what people feel about your brand or your competitors. Our NLP algorithms quickly extract language features like sentiment, significant terms and adjectives to turn customer reviews or tweets into intelligible insights. Drive business impact by modelling the connections between keywords, topics and social communities.

‍

Our Investigations

Transforming Music Industry Analytics: How IDOL and Graphext Empower Artists with Data-Driven Strategies

This article explores how IDOL and Graphext empower artists in the music industry with data-driven strategies, providing tools for audience analysis, campaign performance prediction, and macro-analysis across ad formats and platforms.

Álvaro Otero

Finding the most protein-rich, low-calorie, and healthy foods with data & Graphext.

Advanced analytics enables users to find the most protein-rich, low-calorie, and healthy foods at Mercadona, supporting muscle growth and overall health.

Victoriano Izquierdo

Women's Rise in US Congress and Senate: From Absence to Dominance

Women have transitioned from being virtually non-existent in the US Congress and Senate for almost two centuries to predominating over men in the new generations of politicians.

Victoriano Izquierdo

Turn your Notion database into actionable insights

The Graphext Notion integration allows teams to generate performance reports, user segmentation reports, and analyze user interviews using their Notion database.

Mario Araujo

Why data without profound human interpretation, is meaningless, even in the era of AI

Many people are saying that ChatGPT is going to replace all data analysts and data scientists. Well, that might never happen… Humans bring contextual understanding, identify emerging trends, and build trust in the data analysis process.

Victoriano Izquierdo

A Market Segmentation of 1000 Supermarket Customers Using Data on Sales, Income and Demographics

Our team clustered 1000 supermarket sales in order to segment customers according to their buying habits. Our market segmentation analysis uses data on the demographics, income and geography of customers to identify key buyer personas and inform marketing strategies and campaigns.

Paul Suddon

Clustering countries by Covid-19 Mobility Trends changes

We used the Apple Mobility Trends Report to cluster with Graphext countries that are experiencing similar changes in people's walking behaviour.

Victoriano Izquierdo

Is Mark Cuban a Socialist, a Communist, a Globalist... or Something Else? How Trump Supporters Attack Another Billionaire

Mark Cuban is one of the wealthiest people in America, with an estimated net worth of $4.1 billion. He asked for a tool to work out why his Twitter supporters were calling him a socialist, a communist and a globalist - and to analyze whether their accusations were true!

Victoriano Izquierdo

La imagen que transmite la política española:

Postureo en traje estrechando manos, Baños de masas y selfies, Memes y de relax en la naturaleza. Analizamos las fotos de los 10 políticos españoles más seguidos en Instagram

Maximilian Zachow

How Spanish Politicians and the Media are Connected

In 2017, we analized the Spanish political community and how are they related with the most important media based on how their interactions from their twitter account.

Victoriano Izquierdo

Mastering a Career in Data Science: Takeaways From the 2022 Kaggle Survey

We looked at the insights from the 2022 Kaggle Survey to provide guidance for those interested in pursuing a career in data science or programming, covering topics such as job roles, programming languages, machine learning, and compensation.

Mario Araujo

How we use Graphext to predict who will pay for Graphext with lead scoring

Using Graphext to predict potential customers through lead scoring by selecting the right variables and focusing on engagement with the tool in the near future.

Thomas Buhrmann

Menhir & Graphext- Analyzing the Intangible Value of Financial Assets.

Working at the intersection of data science and finance, Menhir is using Graphext to understand the composition of financial portfolios, performing analysis that typically takes analysts between two and three weeks in just two days.

Andy Clarke

How Aquaservice Use Graphext To Improve Their Prediction Models

We spoke to the data science team at Aquaservice about how they used Graphext to build a clustering model to improve the way they forecast consumer demand. Their project grouped delivery routes using over 30 factors to calculate similarity and exposed patterns in the errors made by their prediction models. Models that are responsible for forecasting the number of water bottles that should be loaded into trucks for delivery across a huge number of routes across Spain.

Victoriano Izquierdo

Overcoming Distribution Challenges with Bicycle Sharing

With Graphext we analyzed the vehicle distribution data (from Kaggle) of a bike sharing service based in Minnesota to try to understand bike shortage in Madrid

Maximilian Zachow

What People Really Feel About Programming Languages

We collected and analyzed answers to a Twitter meme, wherein people were asked to express their relationship with programming languages.

Victoriano Izquierdo

The Method Behind Our Investigation of Reports of Adverse COVID-19 Vaccine Events

Taking on an investigation into the adverse reactions associated with the COVID-19 vaccination rollout in the USA, our team were aware of the increased need for transparency whilst conducting our analysis. This article documents the methodology behind our study of Vaccine Adverse Event Reporting System (VAERS) data.

Andy Clarke

Data-Driven SEO: A Keyword Optimization Guide using Web Scraping & Co-occurrence Analysis (Graphext + Deepnote + Adwords) | Graphext

To improve our SEO, we built a data-driven method to analyze the content of top-ranking Google search results as part of a keyword optimization process. Starting with a single search term, our technique uses web scraping + NLP techniques to find specific keywords that are already proven to boost the rank of similar pages.

Andy Clarke

Finding a New Brand Identity with the Graphext Design Team

As a product evolves, the brand around it evolves too. Here we take you through the design process of finding a new logo for our business, following the experiments carried out by the Graphext design team.

Victoriano Izquierdo

Using Customer Data and RFM Analysis to Create Relevant Ad Campaigns

Drafting in a dataset containing information about purchases made over 4 years by 1590 customers on an online superstore, our team wanted to demonstrate the usefulness of RFM analysis.

Paul Suddon

Finding Real Estate Opportunities in Madrid

We analyzed more than 20k advertisements in real estate websites to try to find underpriced houses with Graphext's predictive algorithms. Along the way we looked into the relationships between prices and factors such as education level or location index to try to find insights and patterns in the data.

Victoriano Izquierdo

Sentiment Analysis & Billboard Top 100: The Changing Mood of Popular Music .

We used sentiment analysis to model 5100 Billboard chart-toppers between 1964 and 2015. Our analysis predicted whether song lyrics were positive, negative or neutral as well as detecting the topic and intent behind the most popular tunes in music history.

Andy Clarke

How We Hacked Private Data for a Social Experiment With Ballantines

We hacked data from people's digital footprints to expose the contradictions between how we perceive ourselves and what our digital behaviour reveals about us. A team from Graphext worked on the social experiment "Me and My Other Me" from Ballantines, excavating and interpreting data concerning people's online activity on digital platforms, the internet and social media.

Andy Clarke

How to Do Exploratory Data Analysis in Python

Python is the most popular language among data scientists for its simplicity and flexibility. Additionally, Python includes multiple libraries to deal with Exploratory Data Analysis (EDA). In this post, we'll perform exploratory data analysis on a dataset using Python code to solve each step of the exploration process.

Alex Gonzalez

ODS y Agenda 2030: Desarrollo sostenible en los medios de comunicación

En este proyecto nos planteamos analizar cuál es el rol que están asumiendo los medios de comunicación en la difusión del desarrollo sostenible y comprender el nivel de sensibilización y concienciación sobre la importancia respecto a la Agenda 2030.

Victoriano Izquierdo

Using Mutual Information to Cluster Variables and Discover the Associations Between Survey Questions

Our team set out to build a type of analysis that could be used to measure the strength of association between variables in a dataset. Here's how we did it ...

Alex Gonzalez

Graphext raised $4.6M to create the best exploratory data analysis platform for building explainable AI models

Graphext has raised $4.6M in a seed round led by Hoxton, bringing their total seed funding to $7M, for their data exploration and predictive modeling platform that enables data scientists to explore data, create predictive models, and collaborate with the business in one place.

Victoriano Izquierdo

Health vs Economy: Using Twitter to Investigate How Latin American Leaders Have Responded to COVID-19

Julián Yunez, a political communications consultant, used Graphext to investigate how messages related to public health and economics have been balanced by Latin American leaders during the pandemic.

Andy Clarke

The Evolution of American Protests After the Death of George Floyd: COVID-19, BLM and the Election

2020 has been a turbulent year for every country but particularly in the USA. We clustered American protest events between May 24 - Nov 28, 2020 to investigate the relationships between types of protests, their violence and their geography.

Andy Clarke

36 Data Podcasts to Follow in 2022

The world of data science podcasting has become as varied as the input parameters to a Linear Regression model. From household names like Freakonomics to less known up-and-comers like Big Data Beard, data professionals are sitting up from their computers to talk about business, the future of AI, data in the real world and much more ... if you know where to look.

Victoriano Izquierdo

Funnel Analysis and Network Visualisation with ING

Working with data from ING, we created a network visualization of how users navigate hundreds of web pages, correlating their navigation patterns with specific user profiles. 

Almudena Sanz

Who to follow to stay up to date with all things data science

A Complete Guide to Remarkable Data-Driven Content: YouTubers, Streamers, Podcasts, and Newsletters You Should Check Out in 2023

Allysa Arocha

Who Will Replace Cristiano Ronaldo in Real Madrid?

Can data tell us who can replace Ronaldo in Real Madrid? Are there any similar players to Cristiano according to data? We analyze 18000 players to find out 

Asier Sarasua

Benchmarking McDonald's Store Performance

Using Graphext, we helped McDonald’s to categorize their more than 500 spanish stores based on their typical customer profiles, to later analyze their sales transaction data.

Almudena Sanz

Las 2 Españas a Palos de Golf: quiénes y cómo son los que reparten tweets a izquierda y derecha.

La semana pasada, millones de personas en España acabaron imaginando en su cabeza algo que realmente nunca pasó: a un señor rico del barrio de Salamanca destrozando mobilario urbano con un palo de golf. Analizamos con datos y Graphext cómo se originó este bulo, y sobre todo quién hay detrás a izquierda y derecha propagando estas nuevas narrativas políticas

Victoriano Izquierdo

Finding the 'Perfect' Sales Candidate Using Clustering and Prediction: Graphext and "The Sales Acceleration Formula"

Exploring how Graphext's data-driven approach might be used to identify the characteristics of successful salespeople.

María López

¿Es la clase obrera española la más xenóbofa de todas las clases sociales? Qué dicen los datos del CIS

Los obreros no cualificados y cualificados tienen un 36% y 16% más de posibilidades de decir que el número de inmigrantes es excesivo, analizamos qué otras variables pueda explicar en por qué tienen esa percepción sobre la población inmigrante en España. 

Victoriano Izquierdo

Campus Madrid Community

During 2017 we were monitoring all the conversations on Twitter about and around Campus Madrid. This is the network of 4838 people who participated in the conversation. 

Victoriano Izquierdo

Exploratory Data Analysis with Graphext vs Jupyter Notebooks: A Comprehensive Guide

In the ever-evolving world of data-driven decision making, the ability to analyze and interpret data is a crucial skill. In this video, I demonstrated how Graphext, our data analysis tool, can be used for exploratory data analysis, even if you're already proficient in Python and Pandas.

Victoriano Izquierdo

The 5 Most Extreme US Office Characters

Testing out our brand spanking new integration with Hugging Face models for NLP, we analyzed speech from characters in all 9 series of the US Office. Added into our Graphext project, the language models focused on classifying the dialogue of Michael, Dwight, Pam, Jim, Daryll and all the other characters according to the detection of sentiment, emotion, offensive language, irony and hate speech.

Andy Clarke

Reverse Engineering Infamous Marketing Strategies from Innocent Drinks

Why are the social media strategies of Innocent Drinks considered as the gold standard for marketing teams the world over? We collected every tweet (10,521) posted by the communication department to deconstruct Innocent's content, style, reach and engagement with a simple topic analysis.

Andy Clarke

Zara vs. Gap: An Instagram Analysis

What are unexpected insights and surprising patterns of the social media strategies of Inditex’s flagship Zara and Gap? In this article, we analyzed their Instagram campaigns since Feb 2017.

Maximilian Zachow

The Moneyball Method: Using Data to Build a Football Dream Team (On a Budget)

Our team set out to build an exceptional football team for less than 100M Euros. Using data provided in the FIFA 2020/2021 dataset - the video game - we built a prediction model in order to find the key performance attributes for each position. Then, we used this to pick out a team of excellent but undervalued players.

Andy Clarke

Las cifras de muertos en España por COVID-19 en contexto.

¿Quiénes son y dónde están los excesos de muertes en España en los últimos años por edad, sexo y Comunidades Autónomas? 

Victoriano Izquierdo

Patriotism, Animals, Comedy and Sex: Clustering 233 Superbowl Ads

We built a model clustering 233 Superbowl ads using data from FiveThirtyEight in order to work out what content brands use to sell their products during America's most-watched sporting event. 

Victoriano Izquierdo

The Lipstick Effect: Did the 2008 Financial Crisis Drive an Increase in Positive Airline Reviews?

We analyzed 30K airlines services reviews and saw that there are clear jumps in ratings marked by the 2008 financial crisis and subsequent economic recovery. Could these factors have impacted consumers, or have airlines improved their services? 

Brais Ramos

Simple Solutions to Prevent Customer Churn

Our team analyzed 7043 current and former customers of a telecoms provider in order to better understand what types of people are most likely to cancel their contracts.

Paul Suddon

19 Data Newsletters to Read in 2022

Newsletters are becoming a popular way to distil news, events and tips as the data landscape becomes busier and busier! These are the kind of emails we love to receive because they help us to stay ahead of the game ... and they are all about data. As well as data newsletters created for business analysts - The Modern Data Stack shares resources, opportunities and tools (we are very proud to have featured) - there are series geared towards data science and AI developments such as The Batch.

Victoriano Izquierdo

Jake's Project: Investigating the Data Behind a Good Day

Andy and María meet with Jake to talk about a dataset he's building about himself. From skating to people he sees to whether he flosses or not - Jake's data offers a unique and deeply personal insight into his life. But what makes the difference between good and bad days?

María López

Understanding Employee Behaviour

Why do people act the way they do? Why do they buy products, quit their jobs, or change partners? Many of these motives are people's behaviors and some can be found in data.

Almudena Sanz

How to Study Brand Conversations with Advanced Text Analysis?

How can we use text analysis of data from Twitter to improve our understanding of markets? This is the question prompting Paul, a strategist in our business team, to scrape tweets about Lloyds bank and conduct a Twitter topic analysis using advanced NLP and network creation. First, he collected tweets using Tractor, Graphext's scraping tool for social media analysis. Then, he analyzed the topics of tweets using network analysis. Here's how he did it ...

Paul Suddon

Make or Break: After 5 Years ... Couples are Less Likely to Break Up

What's the most important milestone in a relationship? According to data from a Stanford study, it's a day like any other that occurs somewhere between the 4th and 5th anniversary of a relationship. 

Victoriano Izquierdo

When Dating Apps Met Survey Theory: Sampling, Weighting & Romance

A picture of a population is what most surveys hope to achieve. Who doesn't want to know which essential Tinder personality traits help a person to be successful in love? We're taking a look at the fundamentals of survey theory - sampling & weighting - through the lens of a Pew Research survey that examines American attitudes towards relationships and dating apps in 2021.

Alex Gonzalez

How Couples Meet and Stay Together014100

Grabbing data from a Stanford University study on relationships, we set out to investigate what a 'happy' relationship looks like and how you find one.

Victoriano Izquierdo

What People Have Felt and Thought About The 2021 UN Climate Change Conference (COP26)

We collected every tweet published about the 2021 UN Climate Change Conference (COP26) to study how people have engaged with events during the summit. Using topic analysis and emotion detection, our project dives into people's visceral reactions to agreements on deforestation, commitments between China and the USA and the appearance of Barack Obama.

Andy Clarke

Good Risk vs Bad Risk: Deconstructing the Features of 1000 German Loans

Attempting to discover the most influential features of a loan application when considering risk, our team built a model using the features of a loan application to predict whether an applicant would have a good or bad risk rating.

Paul Suddon

3 Mapas de Poder Político en España

Hoy publicamos en El País 3 grafos que representan 3 mapas de cómo están conectados en redes sociales los periodistas políticos, los diputados y las cuentas de Twitter políticas más relevantes en España. 

Victoriano Izquierdo

Analyzing a Spanish Election Poll with Aquienvoto

Opinion Poll Company Aquienvoto & Graphext teamed up to look behind the curtains of April 28th’s result. Join us as we analyze how people responded to 45 different questions.

Victoriano Izquierdo

The Top Stories in 2020 according to every tweet from 38 UK news publishers

We collected every tweet in 2020 from 38 UK news organisations to find out what the media have been reporting on. Then we visualised categories of tweets as trends to see what the British media landscape looked like throughout the year.

Andy Clarke

Conspiracies, Complexity and Clustering: Investigating Reports of Adverse COVID-19 Vaccine Effects

Modelling data from the Vaccine Adverse Event Reporting System (VAERS) - a US government-sponsored vaccine reaction monitoring service - our team set out to investigate reports of adverse health effects related to the seismic rollout of the COVID-19 vaccination programme in the USA. 

Paul Suddon

How to Build your Brand Through Social Media

Many companies use Twitter in a traditional way: as promotion, presenting new features. But over 50% of all Tweets have another motivation.

Maximilian Zachow

A Beginners Guide to Market Segmentation: Types, Techniques & Examples to Better Understand Your Customer Base (with Data)

Market segmentation means splitting your customer base into distinct communities based on the similarity of their features. Depending on the data you use to segment customers, clustering a market dataset results in the grouping of customers based on geographic, demographic, behavioural and psychographic factors as well as their buying preferences.

Andy Clarke

How the US Congress Tweeted in 2020

We analyzed almost 200K tweets from members of Congress, comparing across parties and seeing what worked for each one. We found interesting insights by analyzing the way each party worded their opinions. 

Dani Vegara

How to Perform Simple & Effective Customer Segmentation | A Walkthrough with Data from a Delicatessen (Dataset interactive)

Customer segmentation involves splitting a customer base into distinct groups. These customer segments are defined by specific and shared characteristics, behaviours or preferences that help businesses to spot patterns and associate customers with one another. This article walks through the steps involved in a simple customer segmentation analysis. Using sales data from a delicatessen, we'll segment customers according to their buying preferences and behaviour. To achieve this, we'll use a powerful machine learning technique known as clustering. 

Andy Clarke

How to Look Good on Video Calls: Analyzing 1K Skype & Zoom rooms.

How do you look your best in a video call? Collecting almost 1000 reviews of video call backgrounds, we set out to find patterns in what makes people give off good or bad impressions.

Victoriano Izquierdo

Survey Analysis Using Graphext

Surveys are an easy way to gather people’s opinion or behaviour, but the challenge has always been once you have all the data, what do you do with it? 

Almudena Sanz

Nuevas perspectivas en analítica y detección de talento: webinar con D'Anchiano

El pasado 15 de julio estuvimos en directo con Juan Palacios, CEO y fundador de D'Anchiano. Juan nos contó en detalle el uso que le da a Graphext para procesos de selección, así como una mayor visibilidad de qué utilidad tiene Graphext y el análisis de datos en RRHH y detección de talento.

María López

How Data Can Help You Keep Your Workers

To showcase how a company could reduce employee turnover, our team clustered a dataset containing information about IBM employees to discover the reasons why employees left their jobs.

Paul Suddon

Healthy Food: A Tweet Content Analysis

Is healthy food only about being vegan and following diets or is the community mentioning more narratives around that topic? We analyze 30k tweets to find out.

Maximilian Zachow

Social Listening Using Graphext

In this post we will identify and analyze the topics that interest the community of VCs in the US. To accomplish this we will use Graphext and Contexto.

Almudena Sanz

Alexa Amazon Reviews

If you want to succeed with your product in the market you have to listen to what your customers say. Take a look to this post to see how to achieve this with Graphext.

Maximilian Zachow

Graphext | Graphtex | Graphnext: Grouping Similar Spellings Using Chars2Vec and Agglomerative Clustering

'España' and 'Españha' are just spelling variations. We built a way of grouping words spelt differently but referring to the same concept. 

Maria del Mar Ruiz

How We Created a Pandemic-Resistant Team Building Session (Interactive)

Motivated by the prospect of getting to know our new colleagues, our design team created a team-building session that negotiated the limitations brought by the pandemic.

Laura Castro

The Importance of Blockchain to International Development

We recently worked with the United Nations to understand the importance of Blockchain and other emerging technologies in the field of international development. 

Almudena Sanz

22 Data YouTubers + Streamers to Watch in 2022

Youtube, Twitch and other streaming platforms are full of data professionals sharing hacks, tutorials and stories of their working life. As well as content geared towards people starting out with data analysis - like Reuven Lerner covering essential Python tips and walkthroughs - there are videos posted by data Youtubers and streamers that debate topics at the forefront of data science research - Cassie Kozyrkov for instance.

Victoriano Izquierdo

Cómo el Congreso de los Diputados tuiteó durante el Estado de Alarma

Prácticamente todos los diputados del Congreso actual, la XIV legislatura, tienen cuenta en Twitter y gran parte de su trabajo consiste en leer y escribir tweets. Para bien o para mal, los políticos nunca han tenido más poder y facilidad para mandar directamente sus mensajes a los ciudadnos sin el filtro de la prensa. Analizamos las principales narrativas que cada partido movió. 

Victoriano Izquierdo

Project Gallery

@JoeBiden Tweets with Images

Analyzing the visual content posted by Joe Biden on Twitter

Social Media | Shared Images

Netflix Titles: Topic Analysis

NLP based network analysis of the titles published on Netflix.

Text | Topics

Analyzing Disneyland Customer Reviews

Extracting the language features of 30000 Disneyland customer reviews.

Text | Topics

Analyzing Trump's Tweets

Analyzing the topics of Donald Trump's Tweets

Social Media | Topics

Congress Members and Their Tweets

Clustering the topics of tweets posted by congress members

Social Media | Topics

Docs

An Introduction to Natural Language Processing for Text Analysis

In a nutshell, NLP - natural language processing - is computer manipulation of human language. Although scientists have tried to program computers to understand text and speech for more than 50 years, massive recent advancements in the field mean that businesses of all shapes and sizes now use NLP to support decision-making and study markets. NLP encompasses lots of data science tasks but we'll be focusing on text analysis here - roughly defined as the process of deriving insights from written language.

Group Similar Spellings

'España' and 'Españha' are just spelling variations. We built a way of grouping words spelt differently but referring to the same concept.

Technical Docs

Analyzing Reviews

This guide is intended to walk you through the process of analyzing customer reviews with Graphext. We will analyze a dataset of 42,656 reviews about 3 Disneyland branches.

Use Case Guides

Scraping with Tractor

Tractor is a tool to scrape data from popular platforms. Use it to build datasets from information on Google, Twitter and Facebook Ads.

How To

Language Support

Working with text data involves analyzing text in many languages. We support many languages for text analysis including English, Spanish, French, German, Italian, Portuguese, Arabic, Turkish, Basque and Catalan.

Technical Docs

Scraping with Tractor

Tractor is a tool to scrape data from popular platforms. Use it to build datasets from information on Google, Twitter and Facebook Ads.

How To

Language Support

Working with text data involves analyzing text in many languages. We support many languages for text analysis including English, Spanish, French, German, Italian, Portuguese, Arabic, Turkish, Basque and Catalan.

Technical Docs