image_id stringlengths 8 28 | caption_L1 stringlengths 51 785 | caption_L2L3 stringlengths 50 1.21k | chart_type1 stringclasses 11
values | cchart_type2 stringclasses 15
values |
|---|---|---|---|---|
1869.png | This grouped bar chart is titled 'Performace Results on Model Understanding", the y-axis shows participants count, the x-axis shows 3 Described Behaviors Categories. Two model user groups are compared in each described behaviours category. | In the category of Specific words, 11 participants use DeepSeer, 12 participants use LIME. In the Phrase/sentencepattern and State categories, only participants useing DeepSeer are reported. | Bar | Group Bar |
springer_2451 copy 2.png | This boxplot shows the angles for different models. The y-axis shows the Angle in degrees, ranging from 120 to 170 degrees. The x-axis represents different models: Amud 1, Amud 2, Manot, Quneitra, and Qafzeh. For each model there are three aoi conditions: 1, 2, and 3, as indicated in the legend. | The chart shows that all sites have similar median angles, but Amud 1 and Qafzeh exhibit higher dispersion and more extreme values. Amud 2 has the most consistent measurements across aoi groups, while Manot and Quneitra display narrower interquartile ranges with slight shifts across aoi levels. | Boxplot | Grouped Boxplot |
springer_2903.png | This multiple line chart shows the "Nano-Partice Count" on the y-axis in range of 0 to 1.2x10 square 6, against "Time (s)" on the x-axis in range of 0 to 960. Three lines represent different welding test runs: first, second and third. | The plot reveals that the number of tests performed on each subject varies significantly. While the first test run shows a more decreasing trend in the beginning before increasing at around 720 seconds, the second and third run show more fluctuations and an increasing trend. | Line | Multiple Line |
springer_0509.png | This stacked bar chart is titled: 2021 Cohort. It shows numbers from 0 to 100 on the y-axis. The x-axis lists two different categories: 2021 (age3-18) and Final State at 2023. There are three different conditions: UCI, BCI and HAHA. | The chart reveals, that both 2021 and Final state at 2023 have equal numbers. In 2021 the UCI condition had the highest score and in 2023 also. | Bar | Stacked Bar |
Springer_2nd_0787.png | This is a scatter plot showing feature impacts on a GBR model’s output. The x-axis shows SHAP values (impact magnitude) with a scale from -1 to 1, and the y-axis lists 20 model features. Colors represent feature values: blue (low), pink (high). | The plot reveals that recent volatility (RV_LAG1, RV_LAG2) have the strongest influence on model predictions. Higher feature values often push outputs positively, especially for volatility-related features. | Scatter | Simple Scatter |
2096.png | This horizontal bar chart shows the Top 10 Features for Visual Data, the y-axis list 10 features, and the x-axis shows Average Impance on Model Output Magnitude in range of 0.0000 to 0.0020. | Features 'Center ar left heel', and 'center ar right toes', ' center at right middle' show higher impact on model output magnitude, aroun 0.0020. | Bar | Simple Bar |
springer_4295 copy.png | This scatter plot shows the relationship betwen the Number of SLC with substrate enriched and the Rank Percentile. The y-axis shows the Number of SLC with substrate enriched, ranging from 0 to 65. The x-axis shows the Rank Percentile, ranging from 0.0 to 0.3. | The chart shows that as the Rank Percentile increases, the Number of SLC with substrate enriched decrease slightly. | Scatter | Simple Scatter |
arXiv_0870.png | This horizontal bar chart shows random words on the y-axis and "count" on the x-axis ranging from 0 to 25,000. | The chart shows, that "Human" has the highest count with 25,000, followed by "Chicken", "Egg" and "China" all with 11,000. The lowest count has the word "Risk" and "Spreading" with both 3000. | Bar | Simple Bar |
ACM_2906 copy 3.png | This multiple line chart titled HV Polarization shows the relationship between Detection Probability and Observation duration. The y-axis shows detection probability in range of 0.1 to 1.0, and the x-axis shows observation duration in range of 7 to 12. The chart includes five lines representing different heuristic dete... | The plot shows that all four models perform better as the observation duration increases, with the "mentioned Herein" model consistently outperforming the others. The "iForest Detector" and "LOF Detector" models show similar performance, with a slight increase in detection probability as time progresses. The "Fractal D... | Line | Multiple Line |
springer_4121 copy.png | This line chart is titled "PPO - Time taken vs Episode". The x-axis represents the episode number, ranging from 0 to 500, and the y-axis the time taken, ranging from 10 to 40. Three conditions are compared as indicated in the legend: PPO, Failure and Collision. | The chart reveals that there are more failures than collisions. The failures are always at the peak of long times taken and the collisions at short ones. The most common time taken is between 10 and 30. Outstanding long time taken is evenly distributed over all episodes. | Line | Simple Line |
springer_1249.png | This line graph displays the Mean Squared Error (MSE) on the y-axis, ranging from 10⁻³ to 10⁻⁵ on a logarithmic scale, plotted against training epochs on the x-axis, which spans from 1 to 50. The chart compares model training performance across three datasets containing 1,000, 2,500, and 5,000 samples respectively. | Across all sample sizes, the MSE consistently decreases as training progresses, indicating improved model accuracy over time. Notably, the model trained with 5,000 samples (blue line) achieves the lowest MSE, followed by the 2,500-sample set (black line), and lastly the 1,000-sample set (red line), which shows the high... | Line | Multiple Line |
pie_54.png | The chart is a pie chart titled "Most say people’s actions affect spread of COVID-19." It includes a legend with categories: "A great deal," "A fair amount," "Not too much," and "Not at all." Each category is represetned by one slice of the pie. | The majority, 59%, believe actions affect COVID-19 spread "a great deal," while 28% say "a fair amount." Smaller segments indicate less impact, showing a clear trend towards recognizing significant influence. | Pie | Pie |
arXiv_0770.png | This bar chart is titled "Exercise Counts". The x-axis lists different types of exercises, and the y-axis shows the count from 0 to 14. | The chart shows that the most popular exercise is walking, with a count of 15, followed by running with a count of 13. The least popular exercises are dancing, hockey, skiing and tennis with a count of 1. | Bar | Simple Bar |
1399a.png | This bar chart, titled "Trial Time by Technique," shows the mean trial times for six different techniques: Astral (16.37), Chicken (21.77), Grab (24.39), Sliding (17.33), Teleport (15.16), and Throw (24.46). The y-axis represents trial time, and each bar includes an error bar indicating the standard deviation. Signific... | The "throw" and "grab" techniques have the highest average times (24.46 and 24.39 seconds, respectively), while "astral" and "teleport" have the lowest average times (16.37 and 15.16 seconds, respectively). | Bar | Simple Bar |
ACM_2221 copy 4.png | This heatmap chart shows the distribution of the number of nearest used in a typical analysis. The x-axis represents the count of sources, ranging from 0 to 1600, while the x-axis represents a number from 0 to 1000. | The chart shows a distinct peak at around 1400-1600 nearest at around a value of 1000. | Heatmap | Heatmap |
extra_0294.png | This bar chart shows the Number of journal publications on the y-axis from 0 to 1500, and 12 countries are listed on the x-axis, including China, the United States, Canada, etc. | The chart shows that China leads in journal publications with over 1,200, followed by the United States, with over 900 publications. Other countries show many fewer journal publications, around 300. | Bar | Simple Bar |
springer_3222 copy 3.png | This scatter plot is titled 'Comparison with GMPE' and shows the relationship between Joyner Boore Distance on the x-axis and the predicted PGA on the y-axis, ranging from 10^-2 to 10^1. The plot includes three conditions: Observed Data, Random Forest Prediction, and GMPE Prediction, as indicated in the legend. | GMPE predictions follow a smooth decay trend, while Random Forest predictions closely follow observed data. | Scatter | Simple Scatter |
arXiv_0028.png | This multiple line chart plots 'normalized – delta A divide A' value on the y-axis versus Time on the x-axis for different thicknesses levels. The x-axis spans 0 to 300 ps. 10 different thickness levels ranging from 3.3 nm to 14.7 nm are compared, as indicated in the legend. | All conditions show a decreasing trend across time increases. Especially, the conditions 3.3 to 3.9 nm decay rapidly than other conditions. | Line | Multiple Line |
3212b.png | This boxplot shows Average Touch Pressure in Different Conditions. Three intended touch pressure level is shown on the x-axis, touch pressure/g is whon on the x-axis ranging from 1800 to 2800. Two conditions are compared with each touch pressure level: eye-engageed, and eyes-free, as seen in the legend. | Eyes-free condition shows higher touch pressure than the eye-engaged condition. Heacy pressure level has the highest touch pressure compared to other pressure levels. | Boxplot | Grouped Boxplot |
extra_0681.png | The chart is a bar graph titled "Year Level of PSE Delivery." The legend differentiates categories by color, while the x-axis represents year levels (First to Sixth, along with Multiple Years, Longitudinal Y1-Y2, and NS*). The y-axis shows the quantity, ranging from 0 to 8. | The third year level shows a peak in delivery at 7, significantly higher than other levels. This is followed by Sixth (4). First, Fourth and Fifth all show a count of only 1. | bar | Simple Bar |
0225a.png | This bar chart is titled "Distribution of Stress Levels," displays the frequency of different stress levels, ranging from 0 to 9 on the x-axis. The y-axis represents the frequency of occurrences, with values ranging from 0 to 50. | The chart shows a decreasing trend in frequency as stress levels increase. The most common stress level is 0, with a frequency of 48, followed by levels 1 and 2, which have frequencies of 37 and 26, respectively. Higher stress levels (6 to 9) show much lower frequencies, with only 2 occurrences at level 9. This distrib... | Bar | Simple Bar |
ACM_3691 copy 3.png | This multiple line chart shows the "Retention Rate (%)" in Europe on the y-axis, ranging from 0 to 25%, and "Day" on the x-axis, ranging from 1 to 7. The chart includes 7 lines, representing different models and conditions. | The chart shows that the baseline and 1s delay model outperforms other models, with a higher retention rate, especially on day 5 and day 7. | Line | Multiple Line |
ACM_3875.png | This multiple line chart compares the "Comparison of testing set prediction results" between two conditions: "True Data" and "LSTM". The x-axis represents the " Sample Number" ranging from 0 to 250, while the y-axis shows the value of "Ph" ranging from 6.8 to 7.3. | The comparison between the two conditions shows that the "True Data" condition consistently outperforms the "LSTM" condition across all sample numbers. The highest difference occurs at the beginning, where the "True Data" condition has an average difference of 6.85 compared to the "LSTM" condition. As the sample number... | Line | Multiple Line |
springer_4148.png | This scatter plot shows the relationship between the temperature difference and time. The y-axis shows the temperature difference in degrees celsius, ranging from 0 to 6. The x-axis shows the time in minutes, ranging from 0 to 32. | The plot reveals a positive trend between time and temperature difference, with a sharp increase in the temperature difference as time progresses. | Scatter | Simple Scatter |
1320e.png | This bar chart is titled "Computation Time (CT) for Seen Data." It compares the Computation Time of the six models on seen data. The y-axis represents Computation Time (CT) in seconds, ranging from 0 to 25. The x-axis includes three conditions: Indoor, Outdoor, and Quiet. | Computation times for seen data vary slightly across conditions, with a maximum of around 20 seconds, showing efficient processing across different environments. | Bar | Group Bar |
Springer_2nd_0062.png | The chart is a box plot titled "SUS Scores." The y-axis is labeled "SUS Score," ranging from 0 to 120. The plot displays the median, quartiles, and outliers, with a blue box representing the interquartile range and whiskers indicating variability outside the upper and lower quartiles. | The box plot reveals a central tendency around a median SUS score of 50. Outliers go as far as 100 and 0, indicating variability in the data. | Boxplot | Simple Boxplot |
extra_0710.png | This is a box plot titled "Sport and Exercise per Week [Minutes]" showing comparisons between an intervention group and a control group across two time points: Baseline and 12 weeks. The y-axis represents minutes of exercise per week, while the x-axis distinguishes between the two groups. The scale is numerical, indica... | The plot reveals that the intervention group significantly increased weekly exercise minutes from baseline to 12 weeks, contrasting with the control group, which shows minimal change. The median increaed from 0 to 125, approximately. | boxplot | Grouped Boxplot |
extra2_0044.png | The chart is a horizontal box plot. The y-axis presents six categories: Dependability, Stimulation, Efficiency, Perspicuity, Attractiveness, and Novelty. The x-axis is labeled with a scale from 4 to 10, indicating scores. | The box plot reveals notable differences in the metrics, with Perspicuity showing the highest median score closely followed by Novelty, both around 8. | Boxplot | Simple Boxplot |
extra_0092.png | This heatmap is titled "Rho = 100". The x-axis is labeled "Predicted Label," while the y-axis is labeled "True Label". Both axes have the same categories from 0 to 9. A color scale is displayed on the right side, ranging from 0 to 1000, and the cell values represent the count of each category. | The chart shows that the classification results are good. Classes 0 to 4 show strong diagonal dominance with high cell values. Misclassifications are more frequent in classes 5 to 9. | Heatmap | Heatmap |
springer_0674.png | This horizontal scatter plot with error bars is titled "SHAP Values", it shows the SHAP values on the x-axis in range of -15 to 15 against different feature values on the y-axis. For every value, different data points represent SHAP values. | Inflation has a high feature value and the most negative SHAP value of all features. It also shows the highest variance in its data. | Scatter | Simple Scatter |
arXiv_0507.png | This multiple line chart shows the Precision of train and val against epochs. The x-axis represents epochs ranging from 0 to 60, and the y-axis shows Precision values ranging from 0.80 to 0.95. Two lines are displayed: a line for train and a line for val. | The chart shows an increasing trend in Precision as epoch increases for both train and val lines. The train line exhibits a steeper incline, while the val line shows a more fluctuating incline. | Line | Multiple Line |
springer_3951 copy 2.png | This boxplot is titled "B) Lactobacillus"; it shows relative frequency on the y-axis in the range of 0% to 80%. The x-axis lists two Biome conditions: Caa and Mata. Each condition is represented by a boxplot. | The plot shows that Caa Biome has a higher median in relative frequency (40%) than the Mata Biome category. Caa has a larer interquantile range than Mata. | Boxplot | Simple Boxplot |
springer_1959.png | This boxplot shows the Cronbach's alpha value with 95% confidence interval for many different studies. The x-axis shows the alpha value ranging from 0.7 to 0.95, and the y-axis lists the studies. | The boxplot shows that Andreassen et al., Gomes et al., Monacis et al., and Zarate et al. have the highest alpha values of 0.88. The lowest belongs to Chen et al., at 0.73. | Boxplot | Simple Boxplot |
ACM_1779 copy 4.png | This scatter plot titled "B2" shows the relationship between the F1-Score on the x-axis ranging from 0.20 to 0.40 and the AUCROC Score on the y-axis ranging from 0.55 to 0.65. The plot includes four different classifiers: KNN, XGBoost, Random Forest and Decision Tree. each classifier has five datapoints, using five dif... | The plot reveals that all five classifiers perform similarly, with F1-scores ranging from 0.34 to 0.39 and AUCROC Score from 0.57 to 0.64. This is true for all Generators except DistilGPT2, which shows an AUCROC score below 0.53 for all classifiers. The worst performance is that of the XGBoost classifier using DistilGP... | Scatter | Simple Scatter |
1611.png | This bar chart visualizes the accuracy of AdaptiveSound for different sound classes. The y-axis represents accuracy as a percentage, ranging from 0% to 100%, and the x-axis lists various sound classes such as "Microwave," "Baby crying," "Door knock," "Dog bark," and "Speech." Each bar shows the accuracy percentage for ... | The accuracy for the "Hazard alarm" stands out at 98.8%, while the "Doorbell" shows a notably lower accuracy of 88.3%. The highest accuracy is observed for "Microwave" at 99.2%, and the lowest for "Dog bark" at 89.3%, indicating a difference in performance across different sound classes. | Bar | Simple Bar |
ArXiv_3rd_406.png | This bar chart shows the distribution of mentions on the y-axis, ranging from 0 to 10 power of 4, across different categories on the x-axis, labeled as "Artificial Intelligence," "Virtual Reality," "Blockchain," "Robot," "Cloud Computing," "IoT", and '5G'. Each bar represents the number of mentions in each category. | The most mentioned topic is "Robot," with 35,000 mentions, followed by "Blockchain", with 27,000 mentions. The least mentioned topic is "Cloud Computing," with only 1,000 mentions. | Bar | Simple Bar |
extra_0310.png | This group boxplot shows the value of the Daily mean PM10 on the y-axis from 0.0 to 100.0. The x-axis lists four years in the Braganca category: 2019 to 2022. Two conditions are compared within each year category: 'Non-wildfire season months' and 'wildfire season months', as indicated in the legend. | The chart shows that the daily mean PM10 for both conditions are always below 25 from 2019 to 2022. In general, the 'wildfire season months' condition has higher median values in daily PM10 than the 'Non-wildfire season months'. | Boxplot | Grouped Boxplot |
arXiv_0353.png | This line chart shows the MICRO-F1 (y-axis) ranging from 0.930 to 0.970 from May to August (x-axis). Two conditions are compared as seen in the legend: with ontology and without ontology | The chart shows that the "With Onology" group consistently has a higher MICRO F1 than the "Without Onology" group across all months. The difference between the two groups grows over time, particularly between July and August, where the gap is the biggest. | Line | Multiple Line |
pie_154.png | The chart is a pie chart titled "No Party With Edge on Combating Crime, Drugs." It includes a legend with categories: PAN, PRI, PRD, None (Vol), and Other/DK (Vol). Each category is represetend by one slice of the pie. | PAN leads with 28%, followed by PRI at 25%. None (Vol) is 23%, PRD 13%, and Other/DK (Vol) 11%. No party shows a clear advantage, indicating divided public opinion. | Pie | Pie |
Springer_2nd_0336.png | The chart is a scatter plot with regression line. The x-axis labeled "GERD euro per inhabitant - 2019" and ranges from 0 to 2500, and the y-axis labeled "SDG Index - 2023/24" ranging from 55 to 85. | The chart reveals a logarithmic trend, suggesting that as GERD spending per inhabitant increases, the SDG Index also improves, albeit at a diminishing rate. | Scatter | Scatter plot with regression line |
ACM_1649.png | This line chart shows the relationship between the number of features on the x-axis (ranging from 0 to 8) and the accuracy on the y-axis (ranging from 94.5 to 99.0). | The chart shows a general upward trend in accuracy starting at 94.8 accuracy to 98.8 accuracy as the number of features from 1 to 7 increases, with a sharp rise from 1 to 2 features. The line indicates that as the number of features increases, the accuracy also tends to increase, with a slight dip around 5 features. | Line | Simple Line |
extra2_0100.png | The chart is a pie chart titled "Artificial Habitat Type - Coastal Defences." The legend includes four categories: Bars, Breakwater, Seawall, and Shark net, each represented by different segments. | The pie chart demonstrates that Breakwater and Seawall are the most common, while Bars and shark net are barely present. | Pie | Pie |
0064.png | This grouped bar chart shows the Mean and Standard Deviation for Different Word Lengths in 3 Gesture Production: for user-drawn gestures, WordGesture-GAN, and the CLC Model. The word length are categorized into 3 grouped in ranges from 2 to 18 characters on the x-axis, divided into <subranges> 2-4 characters, 5-10 cha... | In the 2-4 character range, the CLC model prediction is closer to the user-drawn gestures, but for the 5-10 and 11-18 character ranges, the CLC model overshoots and is further from the user-drawn gestures than the estimated durations from WordGesture-GAN. | Bar | Group Bar |
arXiv_0039.png | This multiple line chart shows normalized reflectivity R over RF versus Qz for 4 types of Polyethylene Glycol (PEG). The x-axis spans from 0.0 to 0.3 in the reciprocal of Angstroms, and the y-axis has values in the range from 0 to 10. | The four types of Polyethylene Glycol (PEG) follow the similar pattern, they all show some variations around 10 power of 0 reflective values. | Line | Multiple Line |
ArXiv_3rd_328.png | This boxplot is titled 'Bacteria'. The y-axis shows recall scores in the range of 0.4 to 1.0, the x-axis shows six models. Each model is represented by one boxplot. | The plot reveals that the 'Mv1CLT' and 'Mv2CLT' models have a much lower recall median, around 0.75. The other five models 'ST', 'Mv1', 'Mv2', 'Mv1CL', and 'Mv2CL' have higher recall values, all above 0.9. | Boxplot | Simple Boxplot |
springer_0622 copy.png | This line chart is titled 'Topic Mixture Evolution (SMA 3) - part 1' and shows articles % per year over years, ranging from 2000 to 2023. The x-axis shows the years and the y-axis shows the articles % per year, ranging from 0 to 0.5. There are 9 lines representing 9 topics. | The chart shows that topics 1 and 3 start off high in 2000 and decrease over the years. The other topics show somewhat steady growth, while topic 13 shows a larger increase in articles % per year as the years increase. | Line | Multiple Line |
arXiv_0231.png | This horizontal scatter plot with regressin line is titled Effect of Pretraining on simulation dataset for Epidemiology tasks. The y-axis shows 6 models, and the x-axis shows RMSE loss in range of 0.00 to 0.12. 3 conditions are compared as seen in the legend: Brazil COVID-19, Germany COVID-19, and Germany Influenza. | The chart shows that MP-PDE has the lowest RMSE loss for 'Germany COVID 19'. 'RNN' has the lowest RMSE loss for Brazil Covid 19 and Germany Influenza. GraphEncoding has the highest RMSE loss of all methods. | Scatter | Simple Scatter |
extra2_0097.png | The chart is a scatter plot titled "Standardized Residuals vs Fitted Values." The x-axis is labeled "Linear prediction + a[id]" and ranges from 0 to 4, while the y-axis is labeled "Standardized values of (residuals)" ranging from -2 to 4. | The scatter plot reveals a clustering of data around standardized value 0, ranging from 1 to 4 linear prediction, and another cluster at standardized value 2 ranging from 0 to 3 linear prediction. | Scatter | Simple Scatter |
arXiv_1167.png | This line chart shows the "Surivalinal Probability" on the y-axis, ranging from 40% to 100%, and "Time (in days)" on the x-axis, ranging from 0 to 2000. A dashed black reference line is shown in the diagonal in this chart. | Survival probability declines steadily over time, with the rate of decline slowing after about 1500 days. | Line | Simple Line |
springer_1437.png | This mixed chart displays flood data across flood depth ranges (x-axis: 0.1–1.0+ m). Blue vertical bars show the number of flash flood events (y-axis left: 0–200), while three line plots use the right y-axis (0–2000) to represent: (1) sum of displaced people (black line with orange markers), (2) average displaced peopl... | Events peak at the 0.3–0.4 m range, with displaced people and flooded houses following similar trends. | Bar | Simple Bar |
ACM_0181.png | This pie chart is divided into three categories: "Clear and correct," "Not clear or misleading," and "Teacher's exercise." Each category is represetend by one slice of the pie. | The biggest part of the chart is the category "Clear and correct" with 58% followed by "Not clear or misleading" with 33% leaving "Teacher's exercise" at last place with 9%. | Pie | Pie |
springer_0056.png | This grouped bar chart is titled "Per Class Metrics", it shows the Scores on the y-axis in range of 0.0 to 1.0. Six different activities are shown on the x-axis. 4 conditions are compared for each activity, as indicated in the legend: Precision, Recall, F1 Score, and Per-Class Accuracy. | The chart reveals that the highest scores in all conditions are for the categories: Sitting and Standing with all 1.0 scores. Next highest scores has Running. The lowest scores have "upstairs" and "downstairs". The "precision" condition has the highest scores overall. | Bar | Group Bar |
Springer_2nd_0047.png | The chart is a line graph titled "Accuracy." The legend indicates two models: model1 and model2. The x-axis is labeled "Epochs" and ranges from 0 to 120, while the y-axis is labeled "model1 vs model2" and ranges from 0.2 to 1. | The graph illustrates the accuracy trends of both models over epochs, revealing that model1 consistently outperforms model2, especially after a notable dip for model2 after 60 epochs. The overall accuracy stabilization suggests model1’s superior learning efficacy over time, demonstrating a clear performance gap between... | Line | Multiple Line |
ACM_2770 copy 2.png | This line chart is named the "original data". The x-axis shows the Year from 2015 to 2020, and the y-axis shows the Original data from 2 to 18. | The data shows a general upward trend, with the highest value in 2020. The lowest point is in 2015, with a few dips in 2016, 2018 and 2019. | Line | Simple Line |
ArXiv_3rd_630.png | This multiple line chart shows "number of ENVs" on the left y-axis from 0 to 4, and "ENV_0 fraction" on the right y-axis from 0.0 to 1.0. The x-axis shows the changes in Time resolution Delta t. | The number of ENVs remains constant at 2 as time resolution increases. In contrast, the ENV₀ fraction keeps increasing and reaches 0.8 as the time resolution increases. | Line | Multiple Line |
springer_2617 copy 5.png | This bar chart is titled "Busses". The x-axis represents the error, ranging from -4 to 4, and the y-axis shows the number of time intervals, ranging from 0 to 10000. | The distribution shows that most number of time intervals are within the range of -2 to 2 error. | Bar | Simple Bar |
ArXiv_3rd_116.png | This scatter plot with regression line is titled "5p1", it shows SE-tru values on the x-axis in the range of -8 to 6 against SE-pre values on the y-axis in the range of -8 to 6. Two conditions are compared as indicated in the legend: 'sample', and 'y=x' | This plot indicates a positive relationship between the 'SE-pre' value and 'SE-tru' value. The sample scatters are distributed very closely on the line 'y=x'. | Scatter | Scatter plot with regression line |
ACM_1715.png | This grouped bar chart with whiskers shows the QWK for three categories: LLM, Novice, and Expert. The y-axis represents QWK, ranging from 0 to 0.8. The chart compares two conditions: "LLM with feedback" and "LLM without feedback" for each category. | The chart shows that the "Expert" category has the highest QWK. It has a QWK of 0.55 without feedback, and 0.66 with feedback. The whiskers are around 0.1 tall in both directions, both with and without feedback. The other two categories are between 0.45 and 0.47, and the "novice category has whiskers around 0.1 tall in... | Bar | Group Bar |
arXiv_0875.png | This scatter plot is titled "Principal Component Analysis on Unique Natural Image Sets". The x-axis shows Principal Component 1 on a scale from -1 to 1, while the y-axis shows Principal Component 2 on a scale from -1 to 1. Three conditions are compared as indicated in the legend: Intensity, Simulated Events, and True E... | The biggest cluster is at y=0 and x=0. The plot shows that the True Events group has a higher median of principal component 2 compared to the Simulated Events group. The Intensity group has a lower median but a wider spread than the True Events group. | Scatter | Simple Scatter |
ACM_3771.png | This multiple line chart shows the progression of precision on the y-axis in range of 0.000 to 0.035 against the progression of lop on the x-axis in range of 0 to 20. The chart includes 4 lines representing 4 different conditions: "Literature[10]", "Literature (18)", "Literature (19)", and "Ours". | The plot shows that the Precision value declines as the lop value increases. The "Ours" condition consistently outperforms other conditions in precision, and the 'Literature [10]' has the lowest precision value. | Line | Multiple Line |
Springer_2nd_1574.png | This is a heatmap. The x- and y-axes both list 5 variables: from LST_2013 to NDBaI_2013. The color legend ranges from -1 (blue) to 1 (red), indicating correlation strength. | The matrix shows strong negative correlation between NDVI_2013 and NDWI_2013 (-0.43) and strong positive self-correlations (1.0) along the diagonal. Other variable correlations are weak, mostly near zero. | Heatmap | Heatmap |
0212b.png | This boxplot shows the number of rounds bystander participants with four treatments: strict friend, collaborative friend, strict stranger, and collaborative stranger. Number of rounds is shown on the y-axis in range form 0 to 8. | The medians for each group are generally clustered around 3 rounds. The "Collaborative Friend" condition shows a lower interquartile range and a couple of outliers, indicating more variability in performance. The "Strict Friend" and "Collaborative Stranger" conditions exhibit wider spreads, with larger whiskers extendi... | Boxplot | Simple Boxplot |
ACM_2894 copy 3.png | This line chart named VAR shows the Amplitude (V) on the y-axis in range of 100 to 300, against the sample point on the x-axis in range of 0 to 100. | The chart shows that the Amplitude VAR fluctuates significantly across the sample points, with a peak around 20 sample point. | Line | Multiple Line |
Springer_2nd_0370.png | The chart is a grouped bar graph comparing two shapes: Cylinder and Cubic, across two sizes: 50 nanometers and 100 nanometers. The y-axis represents release time points in days, ranging from 0 to 15, and the x-axis shows the groups. | The chart shows that the Cylinder shape released significantly sooner than the Cubic at 50 nanometers, while at 100 nanometers it releaed slightly later. | Bar | Group Bar |
springer_1865 copy 3.png | This line chart shows the yield percentage of seven different bases. The y-axis represents the yield, ranging from 0% to 100%. The x-axis lists the seven bases. | The line chart shows that the Caesium Carbonate base has the highest yield at about 95%, and the potassium acetate has the lowest at about 20%. | Line | Simple Line |
extra_0622.png | The chart is a boxplot showing "log Roach eDNA concentration (gene copy number/L)." It compares two groups, SB2 and SD2, with vertical axes labeled from 0 to 7.5. The legend indicates non-significant results (ns). | The boxplot reveals no significant difference between SB2 and SD2 eDNA concentrations, as indicated by the overlapping medians and the "ns" label. Both medians are around four, with the SB2 median being slightly higher. | Boxplot | Simple Boxplot |
1326.png | This bar chart measures Tak Performance. Five Vestibular Stimulus are shwon on the x-axis: Control, N.GVS, D.GVS, BCV, and CVS. Task Performace Ratings are shwon on the y-axis in rage of 20 to 60. | VCS Vestibular Stinulas has the lowest task performace, aroud 25. N.GVS has the highest task performace ratings, around 40. | Bar | Simple Bar |
springer_2110 copy 2.png | This grouped bar chart shows the relative concentration in percent on the y-axis ranging from 0 to 100. The x-axis shows Time in minutes ranging from 0 to 60. The bars represent two categories: BHMF and HMF, as indicated in the legend. | The grouped bar chart shows that the BHMF category is constantly at 100 percent relative concentration, while HMF decreases over time. | Bar | Group Bar |
extra_0708.png | The chart is a box plot titled "MVPA per day [minutes]", comparing the Intervention and Control groups over Baseline and 12 weeks. The legend distinguishes between the baseline and control groups. The y-axis represents MVPA in minutes, ranging from 0 to 100. | The chart indicates a significant increase in daily MVPA for the Intervention group at 12 weeks compared to Baseline, suggesting effective behavioral change. In contrast, the Control group shows minimal variation, highlighting a trend of greater activity refinement in response to intervention. The means all are around ... | boxplot | Grouped Boxplot |
0968.png | This is a multiple line chart showing the distribution of dark patterns (DPs) across devices. The x-axis represents the count of dark patterns per device, while the y-axis shows the percentage of devices. Two lines are displayed: "Unique DPs" in blue and "Total DPs" in orange, with a legend indicating each line. | The chart reveals that devices with fewer unique dark patterns per device reach a high cumulative percentage (100%) faster than total dark patterns. This indicates that, while the frequency of unique DPs per device is lower, the overall variety of DPs accumulates more quickly across devices compared to the total instan... | Line | Multiple Line |
extra2_0328.png | The chart is a pie chart titled "Main components in Soybean." It includes a legend with various components (Protein, Fat, Carbohydrates, Water, Minerals, Lecithin, Phytates, Isoflavones) and their respective percentages. | The chart reveals that protein constitutes the largest portion at 38%, followed by carbohydrates at 33%, and fat at 19%. Lesser components, such as water (10%) and minerals (5%), are visually distinct, indicating that the soybeans are primarily protein-rich. | Pie | Pie |
ACM_2228 copy 6.png | This multiple line chart is titled "(c)CC with the numbe of servers increase". The x-axis shows the number of servers, ranging from 50 to 150, and the y-axis shows the Communication Cost, ranging from 1 to 1.6. Four lines represent different conditions: GA, GWO, Tabu, and Tabu-GWO. | The chart shows that communication cost remains similar as the number of servers increases. However, the Tabu conditions decreases more than the others. | Line | Multiple Line |
springer_4480 copy.png | This bar chart shows the relationship between eight conditions and Capital Expenditures. The x-axis shows the Capital Expenditures (CAPEX) in euros per kilogram, ranging from 0 to 300'000. The y-axis shows eight conditions. | The chart shows that the Astaxanthin condition has the highest Capital Expenditures and widest range, followed by Carotenoid. Lutein (5%) has the lowest Capital Expenditures, closely followed by Cu-Chlorophyllin (10%). | Bar | Simple Bar |
1922b.png | The multiple line graph displays "Error Rate" on the y-axis and "Total Head Rotation (degrees)" on the x-axis, with control target widths labeled across the top. Three lines represent different head gesture conditions: "Head Gesture (Away)", "Head Gesture (Towards)" , and "Head Gestures (Away + Towards)" with a legend ... | Error rates are highest for small head rotations (around 0.2 to 0.5 degrees) for all gesture types, decreasing significantly as head rotation increases up to approximately 8 degrees. Beyond this point, the error rate stabilizes at a low level for the "Towards" gesture, while it begins to rise again for the "Away" and "... | Line | Multiple Line |
ACM_4335.png | This horizontal bar chart shows the knowledge improvement scores (Cohen's d effect size) for 5 questions for 3 school: India, UK_low, and UK_High. The x-axis represents the knowledge improvement scores (Cohen's d effect size), ranging from 0 to 5, and the y-axis lists the 5 questions: "Overall average - all questions,"... | Overall, the question "What does soap do?" has the lowest Cohen's D effect Scale scores for all three schools. The question "Steps of handwashing" has the highest Cohen's Deficit Scale scores for all three countries. The question "Overall average - all questions" has the highest Cohen's D effect scores for India, while... | Bar | Group Bar |
ArXiv_3rd_721.png | This multiple line is titled 'SVDC UMass coherence scores of all first five sentences'. The x-axis represents the number of topics, ranging from 10 to 200, while the y-axis represents the UMass coherence scores, ranging from -2.5 to 0. Two conditions are compared, as indicated in the legend: 's1' and 's2'. | The plot shows a general decline and very similar trend in UMass coherence scores for both s1 and s2 conditions. They both decreased from -1 to -2 when the number of topics increased from 10 to 200. | Line | Multiple Line |
springer_2900.png | The line chart shows different scenarios on the x-axis: base case, M and N. The y-axis shows the decrease in % ranging from 0 to 30. Two conditions are compared: evacuation time decrease (%) and stress level decrease (%). | The chart shows that both "evacuation Time Decrease" and "Stress Level Decrease" have similar slopes, but differ in their overall behavior. The "evacuation Time Decrease" shows a sharp increase, peaking around 25% after the first scenario, while the "Stress Level Decrease" exhibits a more gradual increase, peaking clos... | Line | Multiple Line |
1924b.png | This box plot, titled "Average Touch Duration in Different Conditions," compares touch duration across three levels: "Long," "Medium," and "Short." The x-axis shows the "Touch Duration Levels," while the y-axis represents "Touch Duration" in milliseconds, ranging from 600 ms to 1600 ms. | Heavy touch condition shows higher rouch duration than Light touch condition in Long and Short levels. In Medoum level, the Light condition shows higher duration time than Heavy condition. | Boxplot | Grouped Boxplot |
springer_1968 copy.png | This line chart shows Monthly Gained Video Views for Adi Hidayat Official. The x-axis shows the date from september 21 to may 24, and the y-axis shows Video Views from 5 million to 20 million. | The line chart shows that the number of monthly video views fluctuates significantly, with several notable peaks, notably around January 22 and April 23. The lowest point is in November. | Line | Simple Line |
0945.png | This multiple line plot is titled "Word Error Rate (WER) of Eavesdroppers at Various Distances across Three Scenes,", it illustrates the Word Error Rate (WER) on the y-axis in range of 0.4 to 1.0, the x-axis shows distance from 20cm to 120cm. 3 types of Scence are compared as seen in the legend: 60dB WER, 50dB WER, an... | The WER increases with distance for each dB environment, with 120 cm consistently presenting the maximum error rate across all scenes. At 60 dB, the highest WER recorded is 100% at 120 cm, while at 40 dB, the least error rate is 47% at 20 cm. This highlights a significant trend where higher sound levels lead to increas... | Line | Multiple Line |
Springer_2nd_1253.png | The chart is a box plot comparing mean temperature anomalies for Fire Years and Non-Fire Years from 2009 to 2022. each year is classifies as either a fire year or a non fire year, so only one boxplot is shown per year. The x-axis represents the year, while the y-axis indicates temperature anomalies. Each box shows the ... | The data reveal that Fire Years consistently exhibit higher temperature anomalies than Non-Fire Years. This trend suggests a significant positive correlation between fire occurrences and elevated temperature anomalies over the observed period. | Boxplot | Simple Boxplot |
ACM_1707.png | This multiple line chart shows the level of engagement across on the y-axis for five clusters based on the "Time Window" on the x-axis, ranging from 10 minutes to 120 minutes. The y-axis has the values of very low, low, medium, high and very high. The chart includes five lines representing different clusters, as indica... | The chart shows that cluster 1 and 2 start at a very high level of engagement, and sharply drops between low and very low after 20 minutes. Clusters 3 and 4 start around medium and drop to between low and very low after 20 to 30 minutes. Cluster 5 starts at low and stays between low and very low the whole time. | Line | Multiple Line |
ArXiv_3rd_740.png | This boxplot is titled "mAP across event encodings - Full Heterogenous models". The x-axis represents four models. The y-axis represents the mAP score from 0.3 to 0.6. | The chart shows that the mAP scores are similar for different models. The TAF model has the lowest mAP median score (0.46). The SHIST model has the highest mAP score (0.48). | Boxplot | Simple Boxplot |
springer_4579 copy.png | This bar chart is titled "Coverage Histogram (0-50X)". The y-axis represents the number of genomic locations, ranging from 0 to 50'000. The x-axis represents the Coverage (X), ranging from 0 to 50. | The chart shows an overall positiv correlation between number of genomic locations and coverage (X). The coverage with the lowest number is 1 and the coverage with the highest number is 47. | Bar | Simple Bar |
ArXiv_3rd_220.png | This boxplot chart shows predicted age (years) on the y-axis, ranging from 0 to 90. The x-axis represents the chronological age (years), ranging from 41 to 91. Each chronological age is represented by a boxplot. A color bar shows the value of MAE for each boxplot; values range from 0 to greater than 10. | The chart shows there is a positive relationship between the predicted age and chronological age. The MAE values are also increasing as the chronological age increases. | Boxplot | Simple Boxplot |
ACM_1768.png | This heatmap shows the correlation ranging from -1 to +1 between two variables measuring different values. | The chart reveals that "tla_synchronous" and "tla_teacher_present" have the highest positive correlation with 0.98, while "tla_onsite" and "tla_online" have the highest negative correlation with -0.18. | Heatmap | Heatmap |
extra_0272.png | This is a heatmap titled "Heatmap of Model Performance by Prompt Type, ITC". The y-axis lists three models: BB, DB, and TI. The x-axis shows two prompt types: General and Specific. Cell values represent performance scores, with a color scale displayed on the right ranging from 0.39 to 0.42. | The heatmap reveals that the 'DB' model performs best overall, and the 'TI' model performs the worst. The 'General' prompt type shows similar performance to the 'Specific' prompt type. The TI model with the specific prompt type has the lowest score, at 0.38, and the DB model with the Specific prompt type has the lowest... | Heatmap | Heatmap |
1931.png | This pie chart is titled "Survey Respondent Vision Level" and represents the vision levels of survey respondents, with categories and percentages for each vision level. The segments include Mild Vision Loss (10.2%), Moderate Loss, Severe Loss, Profound Loss (15.3%), Near Total Blindness (33.9%), Totally Blind (32.2%), ... | The chart shows that the largest group of participants (33.9%) reported near-total blindness, followed closely by those who are totally blind (32.2%). | Pie | Pie |
1109.png | This line graph, titled "WPS Across Age Groups in Dark and Light Modes", shows the Words Per Second (WPS) on the y-axis, ranging from 2.5 to 5.5, and Age groups on the x-axis, categorized as 19 & Below, 20-29, 30-39, 40-49, 50-59, 60-69, and 70+. The chart includes two lines representing Dark Mode (solid black) and Lig... | The graph shows two lines, one for dark mode and one for light mode , both of which descend as age increases. The light mode line consistently remains above the dark mode line, indicating higher wps at all age levels, and the lines do not intersect at any point. | Line | Multiple Line |
extra_0156.png | This heatmap is titled "Mean NO₂ Density (Post-Monsoon)". It shows a map, with Latitude (28°20′N to 28°40′N) on the y-axis, and longitude (77°00′E to 77°20′E) on the x-axis. A legend indicates NO₂ density values in five color-coded intervals, ranging from 0.0010 to 0.0021. | The chart shows that the NO₂ density has spatial heterogeneity with higher concentrations in the east and lower values toward the west. | Heatmap | Heatmap |
springer_0344.png | This grouped bar chart is titled: Distribution of Interaction with VLE. It shows the Frequency on the y-axis from 0 to 50 and the Interaction with VLE on the x-axis from 0 to 100. | The chart shows that the frequency is always above 20. The highest frequency is around 40 and 50 interactions with close to 50 on the y-axis. | Bar | Group Bar |
extra2_0247.png | The chart is a box plot. It has a legend indicating "Covid" and "Non-Covid" as the two groups on the x-axis. The y-axis represents the Alpha-diversity Index: Chao1, with a scale from 400 to 1600. | The chart illustrates that the median Alpha-diversity Index is notably higher in Non-Covid samples compared to Covid ones, at about 1400 compared to 1000, while the covid class has higher variation. | Boxplot | Simple Boxplot |
ACM_1117.png | This heatmap is titled 'Time series heatmap for read accidents'. It plots every month on the x-axis and the hour of day on the y-axis. It shows the number of road accidents for all hours of the day and every month. | The chart reveals no distinct pattern within the hours of day or the month of accidents, except that around hour 19, the number of accidents is above average in general. There is an outlier, month 6 (june) and hour 22, which shows a very high value. | Heatmap | Heatmap |
ArXiv_3rd_420.png | This scatter plot with regression line is titled 'Health', it shows the distribution of shared tweets over time about health, with the x-axis representing the year from 2010 to 2020 and the y-axis representing the share of tweets, ranging from 0 to 1.0. The chart includes three different sentiment categories: Negative,... | The chart reveals that the neutral sentiment dominates, but shows a gradual increase in the share of tweets about health. Both positive sentiment and negative sentiment have lower values and a decreasing trend over the years. | Scatter | Scatter plot with regression line |
Springer_2nd_1532.png | This is a scatter plot. The x-axis shows "DFA Content (Kg/m³)" ranging from 0 to over 200, and the y-axis represents "Compressive Strength (MPa)" ranging from 8 to 22. The legend indicates an R² value of 0.79. | The graph demonstrates a negative correlation between DFA content and compressive strength, as increasing DFA content leads to a consistent decrease in strength, supported by an R² value of 0.79. | Scatter | Simple Scatter |
springer_0111.png | This multiple line chart is not titled. It shows "months" on the x-axis from 0 to 144. The y-axis shows "tumor specific survival (%)" from 0 to 100. Two conditions are being compared: R0 and R1. | The chart shows a decreasing trend for both conditions. R1 has a faster drop from 100% to 0% in 65months. The R0 condition shows a slower decrease going from 100% to 30 in 96months. | Line | Multiple Line |
arXiv_0163.png | This heatmap is titled 'Consistency map (Downsampled) 874mV'. It shows time in seconds on the x axis, ranging from 500 to 4000 seconds. Repeat number is shown on the y-axis, ranging from 0 to 20. The heatmap shows voltage for each combination of x and y. | The map shows a voltage close to 0 for all values, except at time instances just after 1000 sec, 1500 sec, 2000 sec, 2500 sec, 3000 sec and 3500 sec. Then, the voltage increases briefly to 0.20 V, for all repeat numbers, and returns to the basis level of apprximately 0 Volt after. | Heatmap | Heatmap |
springer_3878.png | This multiple line chart shows the cumulative sum on the y-axis ranging from 0.0 to 1.0, against the gray intensity level on the x-axis in the range of 0 to 250. Four lines represent different conditions: Original, CLAHE, Dark Channel Prior, CISACC, as indicated in the legend. | The chart reveals that the cumulative sum of all four conditions increases as the grey intensity level increases. The CLAHE condition shows the steepest increase. The Dark Channel Prior condition exhibits a more linear increase. The CiSACC condition shows the slowest increase with the lowest cumulative sum value. | Line | Multiple Line |
arXiv_0079.png | This multiple line chart is titled 'Reward Comparison Over Time (Correlated)'. The x-axis shows time steps in a range of 0 to 8; the y-axis shows Scaled Reward in the range of 0 to 6. Three conditions are compared as indicated in the legend: No Constraints, Sequential, and Coordinated. | The chart shows the scaled rewards increase as timestep increases for all three groups. The reward comparison shows that participants in the No Constraints group consistently receive higher scaled reward than those in the other two conditions. The scaled reward stabilizes and stops growing at 5 after timestep 4. | Line | Multiple Line |
1179a.png | This grouped box and whisker plot, titled "NASA TLX Responses," shows TLX responses across factors on the x-axis, labeled as Mental Demand, Physical Demand, Temporal Demand, Overall Performance, Effort, and Frustration Level. The y-axis, labeled "TLX Response," ranges from 1 to 7. Each factor compares two conditions: ... | Notable differences include Physical Demand, where the Hand Pointer has a median of around 4, while the Head Pointer has a median of 3. For Effort, the Hand Pointer median is 4 compared to 2 for the Head Pointer, indicating higher perceived demand with the hand pointer. IQRs are between 0 and 2 across these factors. | Boxplot | Grouped Boxplot |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.