Exploratory data analysis and visualisation of datasets

The Exploratory Data Analysis (EDA) tools market is experiencing robust growth, driven by the increasing volume and complexity of data across industries. The rising need for data-driven decision-making, coupled with the expanding adoption of cloud-based analytics solutions, is fueling market expansion. While precise figures for market size and CAGR are not provided, a reasonable estimation, based on the prevalent growth in the broader analytics market and the crucial role of EDA in the data science workflow, would place the 2025 market size at approximately $3 billion, with a projected Compound Annual Growth Rate (CAGR) of 15% through 2033. This growth is segmented across various applications, with large enterprises leading the adoption due to their higher investment capacity and complex data needs. However, SMEs are witnessing rapid growth in EDA tool adoption, driven by the increasing availability of user-friendly and cost-effective solutions. Further segmentation by tool type reveals a strong preference for graphical EDA tools, which offer intuitive visualizations facilitating better data understanding and communication of findings. Geographic regions, such as North America and Europe, currently hold a significant market share, but the Asia-Pacific region shows promising potential for future growth owing to increasing digitalization and data generation. Key restraints to market growth include the need for specialized skills to effectively utilize these tools and the potential for data bias if not handled appropriately. The competitive landscape is dynamic, with both established players like IBM and emerging companies specializing in niche areas vying for market share. Established players benefit from brand recognition and comprehensive enterprise solutions, while specialized vendors provide innovative features and agile development cycles. Open-source options like KNIME and R packages (Rattle, Pandas Profiling) offer cost-effective alternatives, particularly attracting academic institutions and smaller businesses. The ongoing development of advanced analytics functionalities, such as automated machine learning integration within EDA platforms, will be a significant driver of future market growth. Further, the integration of EDA tools within broader data science platforms is streamlining the overall analytical workflow, contributing to increased adoption and reduced complexity. The market's evolution hinges on enhanced user experience, more robust automation features, and seamless integration with other data management and analytics tools.

Unsupervised exploratory data analysis (EDA) is often the first step in understanding complex data sets. While summary statistics are among the most efficient and convenient tools for exploring and describing sets of data, they are often overlooked in EDA. In this paper, we show multiple case studies that compare the performance, including clustering, of a series of summary statistics in EDA. The summary statistics considered here are pattern recognition entropy (PRE), the mean, standard deviation (STD), 1-norm, range, sum of squares (SSQ), and X4, which are compared with principal component analysis (PCA), multivariate curve resolution (MCR), and/or cluster analysis. PRE and the other summary statistics are direct methods for analyzing datathey are not factor-based approaches. To quantify the performance of summary statistics, we use the concept of the “critical pair,” which is employed in chromatography. The data analyzed here come from different analytical methods. Hyperspectral images, including one of a biological material, are also analyzed. In general, PRE outperforms the other summary statistics, especially in image analysis, although a suite of summary statistics is useful in exploring complex data sets. While PRE results were generally comparable to those from PCA and MCR, PRE is easier to apply. For example, there is no need to determine the number of factors that describe a data set. Finally, we introduce the concept of divided spectrum-PRE (DS-PRE) as a new EDA method. DS-PRE increases the discrimination power of PRE. We also show that DS-PRE can be used to provide the inputs for the k-nearest neighbor (kNN) algorithm. We recommend PRE and DS-PRE as rapid new tools for unsupervised EDA.

The three-component reactions of the 16-electron half-sandwich complex CpCo(S2C2B10H10) (Cp = cyclopentadienyl) (1) with ethyl diazoacetate (EDA) and alkynes R1R2 (R1 = Ph, R2 = H; R1 = CO2Me, R2 = H; R1 = R2 = CO2Me; R1 = Fc, R2 = H) at ambient temperature lead to compounds CpCo(S2C2B10H9)(CH2CO2Et) (CHCO2Et)(R1R2) (2–5), CpCo(S2C2B10H9)(CH2CO2Et)(R2–R1–CHCO2Et) (6–9), CpCo(S2C2B10H9)(CH2CO2Et)(CH(Ph)CCHCO2Et) (10), and CpCo(S2C2B10H9)(CH2CO2Et)(CH(Fc)–CH–CCO2Et) (11). In 2–5, one alkyne is stereoselectively inserted into the Co–B bond, one EDA molecule is used to form a sulfide ylide, and the second EDA molecule is inserted into one Co–S bond to form a three-membered metallacyclic ring. At ambient temperature 2–5 undergo rearrangement to 6–9 through migratory insertion of the inserted EDA. Different from 2–5, in 10 phenylacetylene is inserted into the Co–B bond at the terminal carbon and the terminal carbon is coupled with one EDA to afford a six-membered metallacyclic ring with the CO coordination to metal. In 11, a stable Co–B bond is generated, and one EDA and one ethynylferrocene are inserted into the Co–S bond. Moreover, if weakly basic silica is present, 2–4 can lose an apex BH close to the two carbon atoms of o-carborane to give rise to CpCo(S2C2B9H9)(CH2CO2Et)2(R1R2) (12–14) accompanied by the coordination of the two sulfide ylide units to the metal center. The solid-state structures of 2–4, 6–12, and 14 were characterized by X-ray structural analysis.

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

McKinsey Solve Global Assessment Dataset (2018–2025)

🧠 Context

McKinsey's Solve is a gamified problem-solving assessment used globally in the consulting firm’s recruitment process. This dataset simulates assessment results across geographies, education levels, and roles over a 7-year period. It aims to provide deep insights into performance trends, candidate readiness, resume quality, and cognitive task outcomes.

📌 Inspiration & Purpose

Inspired by McKinsey’s real-world assessment framework, this dataset was designed to enable: - Exploratory Data Analysis (EDA) - Recruitment trend analysis - Gamified performance modelling - Dashboard development in Excel / Power BI - Resume and education impact evaluation - Regional performance benchmarking - Data storytelling for portfolio projects

Whether you're building dashboards or training models, this dataset offers practical and relatable data for HR analytics and consulting use cases.

🔍 Dataset Source

• Data generated by Oluwademilade Adeniyi (Demibolt) with the assistance of ChatGPT by OpenAI Structure and logic inspired by McKinsey’s public-facing Solve information, including role categories, game types (Ecosystem, Redrock, Seawolf), education tiers, and global office locations The entire dataset is synthetic and designed for analytical learning, ethical use, and professional development

🧾 Dataset Structure

This dataset includes 4,000 rows and the following columns: - Testtaker ID: Unique identifier - Country / Region: Geographic segmentation - Gender / Age: Demographics - Year: Assessment year (2018–2025) - Highest Level of Education: From high school to PhD / MBA - School or University Attended: Mapped to country and education level - First-generation University Student: Yes/No - Employment Status: Student, Employed, Unemployed - Role Applied For and Department / Interest: Business/tech disciplines - Past Test Taker: Indicates repeat attempts - Prepared with Online Materials: Indicates test prep involvement - Desired Office Location: Mapped to McKinsey's international offices - Ecosystem / Redrock / Seawolf (%): Game performance scores - Time Spent on Each Game (mins) - Total Product Score: Average of the 3 game scores - Process Score: A secondary assessment component - Resume Score: Scored based on education prestige, role fit, and clarity - Total Assessment Score (%): Final decision metric - Status (Pass/Fail): Based on total score ≥ 75%

✅ Why Use This Dataset

• Benchmark educational and regional trends in global assessments
• Build KPI cards, donut charts, histograms, or speedometer visuals
• Train pass/fail classifiers or regression models
• Segment job applicants by role, location, or game behaviour
• Showcase portfolio skills across Excel, SQL, Power BI, Python, or R
• Test dashboards or predictive logic in a business-relevant scenario

💡 Credit & Collaboration

• Data Creator: Oluwademilade Adeniyi (Me) (LinkedIn, Twitter, GitHub, Medium)
• Collaborator: ChatGPT by OpenAI
• Inspired by: McKinsey & Company’s Solve Assessment

EDA Signal Dataset Collected During Startle Events While Walking With a Smart Cane

This dataset accompanies the publication (currently under review):

Villalba-Bravo, R., Grande-Bueno, S., Trujillo-León, A., & Vidal-Verdú, F.
Analysis of EDA signal features under motion artifacts for non-personalized detection of startle events using a smart cane
IEEE SENSORS 2025, Vancouver, Canada.

Description

This dataset includes Electrodermal Activity (EDA) signals collected from seven participants during an experiment in which they walked on a treadmill at a constant speed of 1 km/h while using a smart cane. During the walking task, participants were exposed to auditory startle stimuli designed to elicit stress responses. The smart cane was equipped with a Galvanic Skin Response (GSR) sensor integrated into its handle to continuously record physiological signals in a natural walking context.

The data is organized by participant. All participants provided written informed consent both to take part in the experiment and to allow their anonymized data to be publicly shared for research purposes. Furthermore, the experiment was approved by the Ethical Committee of the Universidad de Málaga (reference 46-2024-H).

Folder Structure

Each folder corresponds to a particiapnt session (e.g., S0/, S2/, etc.) and contains the following files:

S0/
├── S0_DataExperiment.mat
├── S0_audioEventVector.mat
└── S0_SA_Score.mat

...

S8/
├── S8_DataExperiment.mat
├── S8_audioEventVector.mat
└── S8_SA_Score.mat

In addition, the dataset includes a CSV file named caneFeatures_pre_post.csv, containing the extracted features from the GSR, tonic and phasic signals, allowing for the replication of the statistical analyses presented in the study.

File Descriptions

1. S*_DataExperiment.mat

• Description: This file contains the EDA signals acquired at a 4 Hz sampling rate during the experiment, stored in MATLAB .mat format as a structured variable.

• Format: MATLAB Struct (3 fields)

  • GSR: Contains the raw GSR signal along with associated time information: TimeStampDate (UTC date-time format) and TimeStampPosix (POSIX timestamp).

  • TONIC: Contains the tonic component of the EDA signal with the same timestamp fields.

  • PHASIC: Contains the phasic component of the EDA signal with the corresponding timestamps.

2. S*_audioEventVector.mat

• Description: This file contains information about the timing of the auditory startle stimuli presented during the experiment. The data is stored as a MATLAB struct sampled at 32 Hz.

• Format: MATLAB Struct (3 fields)

  • data: A binary step signal indicating the presence of auditory events (0 = no stimulus, 1 = stimulus being played).

  • TimeStampDate: A vector of timestamps in MATLAB datetime format, corresponding to each sample in the data field.

3. S*_SA_Score.mat

• Description: This file contains the self-reported State Anxiety (STAI-State) scores provided by each participant before and after the experimental session. The data is stored as a MATLAB struct.

• Format: MATLAB Struct (2 fields)

  • Training: Numeric score reported after the training session.

  • Experiment: Numeric score reported after the experimental session.

Contact Information

For any questions or further information regarding this dataset, please contact fvidal@uma.es.

Context

Every story has a question that triggered it. Mine was - What are the vaccinations being administered in USA? What are people's reported incidents post the vaccine doses ?

I look at awe at which folks do EDA in kaggle and I have a long way to go.But I want to start small and I have already started my journey.The folks who do wonderful EDA are my source of inspiration and I learn by doing their notebooks in R. Iam a R fan for now.

Content

The data here was downloaded on 29th March from CDC Wonder site which helps take reports on VAERS.

Acknowledgements

My google search on VAERS and Kaggle search for VAERS got me a wonderful notebook and dataset. Thanks to folks like Ayush Garg and jmreuter for helping folks like me learn more.

Inspiration

What are the vaccinations being administered in USA? What are people's reported incidents post the vaccine doses ? Which vaccine has most side effects in all age groups ? Which vaccine has most side effects in each state?

Context

Tailor made data to apply the machine learning models on the dataset. Where the newcomers can easily perform their EDA.

The data consists of all the features of the four wheelers available in the market in 1985. We need to predict the **price of the car ** using Linear Regression or PCA or SVM-R etc.,

Predavanje za predmet Tehnike obrade biomedicinskih signala na master akademskim studijama na Elektrotehničkom fakultetu Univerziteta u Beogradu.

This dataset includes skin conductance response (SCR) measurement, keypress responses and keypress response times to stimuli drawn from the International Affective Picture System for each of 24 healthy unmedicated participants (12 males and 12 females aged 27+/-4.6 years). The experiment used a 2x3 factorial design with the factors picture type (aversive, neutral), and mean ISI (3s, 9s, and 19s).

Data are untrimmed. The referenced article used a trimmed version of the data (trim points: 0.5 s before first marker until 20 s after last marker). This detail is not mentioned in the methods section of the paper.

See the readme file for more detail. Data are stored as .mat files for use with MATLAB in a format readable by the PsPM toolbox (pspm.sourceforge.net).

Prior to statistical analysis of mass spectrometry (MS) data, quality control (QC) of the identified biomolecule peak intensities is imperative for reducing process-based sources of variation and extreme biological outliers. Without this step, statistical results can be biased. Additionally, liquid chromatography–MS proteomics data present inherent challenges due to large amounts of missing data that require special consideration during statistical analysis. While a number of R packages exist to address these challenges individually, there is no single R package that addresses all of them. We present pmartR, an open-source R package, for QC (filtering and normalization), exploratory data analysis (EDA), visualization, and statistical analysis robust to missing data. Example analysis using proteomics data from a mouse study comparing smoke exposure to control demonstrates the core functionality of the package and highlights the capabilities for handling missing data. In particular, using a combined quantitative and qualitative statistical test, 19 proteins whose statistical significance would have been missed by a quantitative test alone were identified. The pmartR package provides a single software tool for QC, EDA, and statistical comparisons of MS data that is robust to missing data and includes numerous visualization capabilities.

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

This event has been computationally inferred from an event that has been demonstrated in another species.

The inference is based on the homology mapping from PANTHER. Briefly, reactions for which all involved PhysicalEntities (in input, output and catalyst) have a mapped orthologue/paralogue (for complexes at least 75% of components must have a mapping) are inferred to the other species. High level events are also inferred for these events to allow for easier navigation.

More details and caveats of the event inference in Reactome. For details on PANTHER see also: http://www.pantherdb.org/about.jsp

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

Defence Data is collected by the European Defence Agency (EDA) on an annual basis. The Ministries of Defence of the Agency’s 27 participating Member States (all EU Member States except Denmark) provide the data. EDA acts as the custodian of the data and publishes the aggregated figures in this booklet. 2012 data does not include Croatia which became the 27th EDA Member State on 1 July 2013.

The data are broken down, based on a list of indicators approved by the Agency’s Ministerial Steering Board. This list has four sections, represented in the headings of the booklet:

General: macro-economic data to show how defence budgets relate to GDP and overall government spending.

Reform: major categories of defence budget spending – personnel; investment, including R and T operation and maintenance and others – to show what defence budgets are spent on.

European collaboration: for defence equipment procurement and R and T to show to what extent the Agency’s pMS are investing together.

Deployability: military deployed in crisis management operations to show the ratio between deployments and total military personnel.

The definitions used for the gathering of the data and some general caveats are listed at the end of the brochure.

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.

Subscribers can find out export and import data of 23 countries by HS code or product’s name. This demo is helpful for market analysis.