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 |
|---|---|---|---|---|
Springer_2nd_0182.png | The chart is a scatter plot plotting CS (MPa) vs. Dry Density (kg/m3). It includes a dashed regression line and shows the relationship between CS on the vertical axis ranging from 0 to 4, and dry density (kg/m³) on the horizontal axis, ranging from 1400 to 1800. | The chart indicates a weak correlation (R² = 0.12) between dry density and compressive strength, suggesting minimal influence of density on strength. While CS generally increases with dry density, the scattered data points suggest substantial variability. | Scatter | Scatter plot with regression line |
1870a.png | This pie chart represents system accuracy, displaying the proportion of correctly displayed predictions versus missed predictions. The chart shows that 84.4% of the predictions were correctly displayed, while 15.6% were missed. | Most of the system's predictions were accurate, with only a small portion categorized as missed. | Pie | Pie |
springer_3100 copy 3.png | This grouped bar chart shows the global warmin potential (Mt Co2-Eq) over time for three models: LGS-LBT-GR, MGS-MBT-GR, and HGS-HBT-RR, with the x-axis representing the years from 2024 to 2040. The y-axis represents the global warming potential, ranging from -300 to 0.0. Each model is represented by a different color,... | All three strategies show increasing global warming potential over time. The HGS–HBT–RR (dark blue) strategy results in the largest global warming potential, reaching below -250 by 2040. The MGS–MBT–GR (light blue) shows moderate impact, while LGS–LBT–GR (orange) has the least but still significant reduction. | Bar | Stacked Bar |
ACM_2281 copy.png | This bar chart shows the probability distribution of 4096 shots on computational basis state. The x-axis represents different computational basis state, and the y-axis shows the probability, ranging from 0 to 1. | The chart reveals that the Computational Basis State is highest at 00000 and lowest at 00110. | Bar | Simple Bar |
Springer_3rd_053.png | This heatmap is titled 'Confusion Matrix'. The x-axis represents the predicted label, and the y-axis represents the True label. Both the x- and y-axis have the same 15 color categories. The cell value represents the percentage of classification, ranging from 0.0 to 1.0. | The chart shows diagonal values are high (e.g., purple 98.3%, yellow 97.5%), indicating strong classification. Lower accuracy appears for beige (66.0%) and tan (66.3%). A notable misclassification happens between white and silver. | Heatmap | Heatmap |
springer_4585 copy 2.png | This bar chart shows the relationship between the probability of target attainment and two categories. The y-axis represents the probability of target attainment - PTA in percent ranging from 0 to 100. The x-axis represents the two categories: 1 x MIC and 4 x MIC. Nine target attainment categories are compared as indic... | The chart shows that the probability of target attainment across all conditions is higher for 1x MIC then 4 x MIC. The condition Ulldemolins has the highest score for both categories and Gregoire the lowest. | Bar | Group Bar |
ArXiv_3rd_806.png | This heatmap is titled “3m Threshold” with predicted labels on the x-axis and true labels on the y-axis. Both axes have two categories of labels. The x-axis shows two categories: 'pred>t' and 'pred<t'. The y-axis shows two categories: 'actual>t' and 'actual<t'. Cell values represent percentage, and a color bar that sca... | Most predictions are correct, with 20.42% for "actual > t" and "pred > t", 70.81% prediction for "actual < t" and "pred < t". There are a few incorrect predictions, 4.74% wrong predictions for "actual < t" and"pred > t", and 4.03% wrong predictions for "actual > t" and"pred < t". | Heatmap | Heatmap |
arXiv_1025.png | This is a scatter plot with a color gradient encoding from 0.0 to 1.0, indicated by a vertical colorbar. The x-axis is labeled "Tgt Superpoints NC" and the y-axis is "Src Superpoints NC," both ranging from 0 to 600. | The plot reveals a dense concentration of points in the lower-left region, gradually tapering off toward the bottom right. This downward trend suggests an inverse relationship between Tgt and Src superpoints NC. | Scatter | Simple Scatter |
springer_1767.png | This line chart plots efficiency (y-axis from 0 to 1.2) against GPU count (x-axis: 2⁸ to 2¹¹). It compares three model sizes: SMALL, MEDIUM, and LARGE. | All models start near full efficiency at 2⁸ GPUs. As GPU count increases, efficiency declines, especially for the SMALL model, dropping below 0.7 at 2¹¹. MEDIUM and LARGE maintain higher efficiency (0.95 and 0.85 respectively), indicating better scalability. | Line | Multiple Line |
springer_0673 copy 2.png | This horizontal bar chart shows the importance of various features. The y-axis shows different economic features, the x-axis shows the importance in percentage, ranging from 0 to 20% | The most important feature is inflation with 13.15%, followed by trade openness with 9.49%. The least important feature is 'Patent stock', with only 3.86%. | Bar | Simple Bar |
extra2_0242.png | The chart is a box plot titled "Chao1 Richness Estimator." It compares two groups on the x-axis: COVID and Non-COVID. The y-axis is labeled "Chao1." The scale on the y-axis ranges from 800 to 1300. | The COVID group has a lower median around 900 compared to the Non-COVID group at approximately 1150. The interquartile range is wider for the COVID group, suggesting greater variability. | Boxplot | Simple Boxplot |
Springer_2nd_1777.png | The chart is a grouped bar graph showing Page Length Comparison in Trials. It includes three datasets (Clinical trials, Non-clinical trials, and All studies) and shows the counts for page lengths across two categories: ≤3 pages, >3 pages, along with summary statistics (Median, Maximum, Minimum). The vertical axis measu... | The chart reveals that Non-clinical trials and all studies consist of fewer than three pages for all three study categories. Clinical trials are more often longer than three pages, with a median of 6, indicating that these studies typicaly consist of more pages. | Bar | Group Bar |
springer_1564 copy 3.png | This boxplot presents total nitrogen (TN) concentrations in mg/kg for CK, PBRS, and HuaYS treatments. The y-axis ranges from approximately 900 to 1900. | HuaYS has the highest median TN and the tightest interquartile range, indicating high and consistent nitrogen content. PBRS follows with a moderate spread, while CK shows the lowest median and widest variability. Statistical labels (a, b, c) suggest that all three treatments differ significantly in TN levels, with HuaY... | Boxplot | Simple Boxplot |
springer_1964.png | This scatter plot shows the relationship between the standard error and the Fisher's z transformed correlation coefficient of each data point. The y-axis represents the standard error, ranging from 0.057 to 0, and the x-axis represents z value ranging from 0.15 to 0.45. | The scatter plot shows that the lowest error achieved is around 0.006, with z value of about 0.2. The highest is about 0.057 with z value of about 0.18. | Scatter | Simple Scatter |
springer_3213.png | This grouped bar chart shows 'Porosity' on the y-axis, ranging from 0% to 10%, against 'Curing time (days)' on the x-axis, ranging from 0 to 150. The chart compares six models: 0FA, 35FA, 50FA, 35GGBS, 65FA, and 10SF. | The chart shows that the '0FA' model outperforms the other models across all time periods, with the highest cumulative probability of success during the first 7 days, followed by a decline in performance in the subsequent days. | Bar | Group Bar |
springer_3220 copy 4.png | This scatter plot is titled 'Predicted vs. Actual Values for Arias Intensity'. The y-axis shows Actual Arias Intensity in the range of 10 to the power of -5 to power of 2 meters per second square, the x-axis shows Predicted Arias Intensity (m/s) in the range of 10 to the power of -6 to power of 2 meters per second squa... | The plot reveals a positive trend between the predicted and actual arias intensity values. The training data and testing data follow the similar trend. | Scatter | Simple Scatter |
2528.png | This bar chart is titled 'SUS Score of 4 categories', it has SUS Score on the y-axis ranign from 0 to 100, 4 Categpries are listed on the x-axis. | The category 'Wall+Tabletop' has the highest SUS score. abpve 80, and 'Body-fixed +Floor' has the lowest SUS score, around 60. | Bar | Simple Bar |
arXiv_1055.png | This pie chart shows the distribution of the OD Population growth across 4 categories. The chart is divided into 4 equal sections, each representing a category. | The chart shows that 66.7% of the population growth falls under "1" category, followed by "3" and "2" categories.. | Pie | Pie |
springer_1077.png | This horizontal stacked bar chart represents the opinions of participants on the question "Suicides should not be reported..." The y-axis lists three scenarios: ".. to respect the privacy of family and friends and not exacerbate possible feelings of guilt," "... becausethey can trigger other people to die by suicide," ... | The majority of people agree with responses 'to repsect family and friends' (60%) and 'because they can trigger other suicides' (50%). More people disagree with 'it's only sensationalist interest' (40%). | Bar | Stacked Bar |
Springer_2nd_1155.png | The chart is a line and scatter plot showing "Π Insurer vs. σ". The legend indicates two series: OBP and UP. The x-axis represents σ (sigma) and ranges from 0 to 250, while the y-axis shows Π Insurer ranging from 0 to approximately 4500. | The chart illustrates OBP having a very stable level of around 1100 with increasing σ, while UP starts off at levels above 4000 and makes a very sudden dive to 200 as sigma attains a value of apprximately 60. This constrast indicates large differences in the behavior of OBP and UP for different values of sigma. | Scatter | Scatter plot with regression line |
Springer_2nd_1222.png | This heatmap has predicted labels on the x-axis and true labels on the y-axis. Both axes have the same two categories of labels: 'Bad' and 'Good'. Cell values represent a count of classification. | The chart shows that the classification for the two classes is highly accurate. For the 'Bad' class, 925 are correct, and 3 are wrongly classified. For the 'Good' class, 984 are correctly classified, and 13 are wrongly classified. | Heatmap | Heatmap |
springer_0340.png | This line chart shows the accuracy over epochs. The x-axis represents epochs from 0 to 100 and the y-axis represents accuracy, ranging from 0 to 1. The line is titled: AffectNet. | The chart shows a increasing trend in accuracy over the epochs with a bit of fluctuation. It starts at 0.2 and ends at 1.0 at epoch 100. | Line | Simple Line |
springer_3977.png | This bar chart is titled "Rainfall vs. Crops". The y-axis shows different types of crops, and the x-axis shows values (Measureure) ranging from 0 to 200. | The chart shows that rice has the highest rainfall value (over 200), followed by coconut. Muskmelon has the lowest rainfall values, around 25. | Bar | Simple Bar |
ACM_1164 copy.png | This is a heatmap chart titled "right hand". The y-axis shows frequency ranges in hertz (Hz), starting at 4–8 hertz up to 36–40 hertz. The x-axis is labeled "frequency band index," with values from 1 to 9. There is a color shaded scale on the right with represented valuew from 40 to 110. | The most notable signal intensities are seen in: 8–12 hertz, index 3; 12–16 hertz, index 6; 20–24 hertz, index 1 and 5; 24–28 hertz, index 2; 28–32 hertz, index 4; 32–36 hertz, index 5 and 24–28 hertz, index 9. | Heatmap | Heatmap |
springer_2656.png | This grouped bar chart shows the distribution of Decision Tree metrics for different metrics. The x-axis lists metrics: Accuracy, Precision, Recall, F1, Fbeta, and AUC. The y-axis represents the metric values. | The chart reveals that Precision has the highest values at 99.48, and Recall has the lowest at 98.91. | Bar | Group Bar |
arXiv_0737.png | This chart shows the Value on the y-axis ranging from 2-12 and the x-axis lists the parameters such as λ lambda, μ mu, and K, marked by red dots, with vertical black lines showing variability. | The chart shows values mostly range from 3 to 10. Parameters like μ5, μ6, and K13 have higher means, while K2 and K23 are lower with tighter error margins, indicating greater consistency. Larger error bars, especially for μ4 and μ5, reflect more uncertainty. The chart conveys both central trends and reliability across ... | Scatter | Scatter plot with error bars |
ACM_3320.png | This multiple line chart shows System Concurrency Processing Capability. The x-axis shows Concurrent Task Count, ranging from 0 to 100, while the y-axis shows Processing Capability (tasks/sec). Two conditions are compared: the Traditional System and the Containerized System, as indicated in the legend. | The chart shows that both systems exhibit a positive trend, with the "Containerized System" consistently outperforming the "Traditional System" across all task counts. The gap between the two systems increases as the task count increases, indicating that containerized systems are more capable of handling larger workloa... | Line | Multiple Line |
ACM_2221 copy 3.png | This heatmap chart shows the distribution of the number of bicubic 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 bicubic at around a value of 1000. | Heatmap | Heatmap |
0292.png | This boxplot, labeled "(a) Image Classification," presents NASA-TLX scores across six conditions: control, god, human, animal, plant, and object. The x-axis represents the conditions, while the y-axis, labeled "NASA-TLX score," ranges from 0 to 100. Each box plot shows the median, interquartile range, and individual ou... | The chart reveals that the "control" condition has the lowest NASA-TLX scores, with minimal spread and several low outliers, suggesting lower task load. In contrast, the "god" and "human" conditions have higher median scores and greater variability, indicating increased cognitive demands. "Animal," "plant," and "object... | Boxplot | Simple Boxplot |
1925b.png | This bar chart displays a comparison of word count between the control and experiment (Captivate!) groups. The y-axis represents word count, ranging from 0 to 400, while the x-axis shows the two groups: Control and Captivate!. The control group averaged 308 words, whereas the experiment group averaged 301 words, with e... | This comparison reveals a slight difference in average word count between the two groups, with the control group having a marginally higher word count. | Bar | Simple Bar |
0531.png | This grouped bar graphis titled 'performance of four models' based on their Avg. accuracies on the y-axis across 3 conditions on the x-axis: all sounds, Med. Priority and High Priority. Four types of mmodels are compared within each conditions: VGG-lite, ResNet-lite, Inception, and MobileNet. | VGG-lite model has the highest avg.accuracy across all conditions, followed by ResNet-lite, Inception, and Mobilenet. HIgh Priority condition has higher avg.accuracy than other two conditons. | Bar | Group Bar |
ACM_1404.png | This is a vertical bar chart titled "Counts by Relation Type." The X-axis shows seven relation types, and the Y-axis represents count values ranging from 0 to 3,000. A legend labels the count. | The "HasContent" relation type has the highest count at 3,085, while "HasSubText" has the lowest at 219. Most relation types have counts below 1,000. | Bar | Simple Bar |
ACM_1661.png | This boxplot shows the average preferences for feedback by AI across four classes. The y-axis represents the average preference, ranging from 1 to 5, and the x-axis lists the four classes: Class A, Class B, Class C, and Class D. | The boxplot shows that the average preferences for feedback by AI has the highest value for Class A with 4.34 while the median is around 2.75 and therefore the lowest of all classes. Overall the median values for all 4 classes are similar, with values ranging from around 2.75 to around 3.75 which were class B and Class... | Boxplot | Simple Boxplot |
1473b.png | This scatter plot, labeled as (b), shows session clusters based on app-switching frequency. The x-axis represents the binary logarithm of the app switching frequency, with a range of 0 to 0.9, while the y-axis remains the binary logarithm of the text-change frequency, with a range of 0 to 1, as in subfigure (a). | In this plot, cluster D shows clear separation from all other clusters based on its high application switching frequency, unlike in subfigure (a), where cluster D was mixed with the other features. | Scatter | Simple Scatter |
extra_0385.png | This pie chart shows the breakdown of two categories: Inland and Coastal. Each category is represetned by one segment of the pie, with amount and percentage labelled. | The chart shows that the Inland areas dominate with 97.5% coverage (124,110 Ha), while coastal regions contribute only 2.4% (3,035 Ha). | Pie | Pie |
extra_0465.png | This heatmap is titled 'Confusion Matrix 2015'. The y-axis shows True labels, and the x-axis shows predicted Labels. Both the x- and y-axis have the same six categories, including Agric, BuiltUp, DensVeg, Water, BareSoil, and MxdGrass. The cell values represent the count of classification, ranging from 0 to 5000. | MxdGrass (5278), Water (1019), and BuiltUp (916) show strong diagonal dominance, indicating high prediction accuracy. There is still some moderate confusion with Agric and MxdGrass. | Heatmap | Heatmap |
springer_4399.png | This scatter plot is titled "cor=0.29, p=1.4e-28". The x-axis shows Membership in Grey in range of 0.0 to 0.6, the y-axis shows Significance for Cancer in range of 0.0 to 0.6. | The plot reveals that the higher the membership in grey the greater the density of data. | Scatter | Simple Scatter |
Springer_2nd_1763.png | This is a Multiple Line chart comparing correctness rates across eight questions (Q1–Q8) for three categories: Oral Medicine, Non-Oral Medicine Medical, and Non-Medical. The x-axis lists questions Q1 to Q8, while the y-axis shows correctness rates from 0.0 to 1.0. | Oral Medicine consistently scores the highest, peaking near 0.95 at Q6. Non-Oral Medicine follows closely, while Non-Medical trails behind, especially at Q2 and Q7 with the lowest rates around 0.15 and 0.35. The chart shows clear performance gaps between groups. | Line | Multiple Line |
ArXiv_3rd_281.png | This horizontal stacked bar chart shows the distribution of damage types on a scale from 0 to 100%( x-axis) across four different conditions (y-axis): B-10KeV, B-30KeV, As-10KeV, and As-30KeV. Each bar is divided into segments of different damage types. | The chart reveals that the B-10keV condition yields the most balanced distribution. As-10keV has 83% surface damage and only 3% undamaged. As-10keV condition has 44% of undamaged. | Bar | Stacked Bar |
springer_2996.png | This horizontal bar chart shows the percentage of underrepresented minorities in internal medicine sub-specialties. The y-axis lists the 10 specialities (geriatric medicine, infectious disease, rheumatology, endocrinology, hematology &oncology, nephrology, pulmology &critical care, gastroenterology and cardiology), and... | Geriatric medicine has the highest percentage of underrepresented minorities with 13.9%, followed by infectious diseases with 12.8% and then neprology at 10.7%. The lowest percentage is reached by pulmonology at 6.3%. | Bar | Simple Bar |
springer_3786 copy.png | This horizontal bar chart shows the Climate Change Impact on different agricultural systems. The y-axis lists 5 types of agricultural systems, and the x-axis shows the scale from 0 to 100. | The chart indicates that vertical flow constructed wetland and water storage have the highest impact, scoring 65 and 65, respectively. Conversely, decertified composting has the lowest impact, scoring only 38. | Bar | Simple Bar |
springer_0562 copy.png | This bar chart shows the Dose of FWH in ppm from 0 to 160 on the x-axis. The y-axis shows the Microbial load * 10^7 CFU/g of gut, on a range from 0 to 7.5. | The chart shows a increase at first, with a peak at over 7.5 in dose 10. After that the loads drop again and then range around 4 to 5. | Bar | Simple Bar |
ACM_1971.png | This bar chart titled "Summative underperformers" shows the Mastery Probability on the y-axis ranging between 0.00 and 1.00 for several values. | The chart shows that F_LO_5 has the highest Mastery Probability of around 0.90. The lowest Mastery Probability belongs to S_LO_3, which is around 0.15. | Bar | Simple Bar |
Springer_2nd_1265.png | The chart is a bar graph titled "Proportion (%)" showing the substrate preferences of *P. barberinus*. It features two legends representing the "Number of fish individuals" and "Substrate availability". The y-axis is labeled "Substrate", and the x-axis shows percentage with a range from 0 to 100%. | The chart reveals that sand has the highest number of fish individuals (27 out of 30). The highest substrate availability is observed for 'Live corals' with close to 40% of the available substrate. | Bar | Group Bar |
Springer_2nd_1530.png | This is a scatter plot titled “Scatter Plot B: Sand Content (Kg/m³) vs Compressive Strength (MPa).” The x-axis shows sand content from 500 to 700 Kg/m³, and the y-axis shows compressive strength from 8 to 22 MPa. The legend displays an R² value of 0.79. | The scatter plot shows a positive correlation between sand content and compressive strength. As sand content increases, compressive strength generally rises, supported by a strong R² value of 0.79. | Scatter | Simple Scatter |
ACM_1147.png | This is a scatter plot. The x-axis and y-axis represent two principal components ranging approximately from -10 to 15. The legend shows red for Colon_Cancer and blue for Thyroid_Cancer. | The scatter plot reveals two distinct clusters: red dots (Colon_Cancer) concentrated on the left, and blue dots (Thyroid_Cancer) on the right, indicating clear separation between cancer types. | Scatter | Simple Scatter |
extra_0652.png | The chart is a color-coded map titled "O_3 2011-2020 average data map." It uses a gradient legend indicating average ozone levels in µg/m3, with the x-axis representing longitude and the y-axis representing latitude. The scale ranges from 56 to 120. | The map reveals a spatial distribution of average ozone levels across China from 2011 to 2020, with higher levels along the eastern bayside urban region and the northern part of China. Lower levels are shown on the northeastern and southwestern sides, highlighting the difference in ozone concentrations. | heatmap | heatmap |
1835.png | This bar chart is titled Mean Production Times for iFAD Gesture Categories. The y-axis shows Production time [s] in range of 0 to 3.0, the x-axis list 4 iFAD Gesture Categories. | Arm-level has the highest mean production time, with 2.14s. Hand-level shows the lowest mean production time, with 1.6s. | Bar | Simple Bar |
Springer_2nd_0288.png | The chart is a vertical bar graph titled "Wildfires." It depicts wildfires over time from 1900 to 2020, with the vertical axis ranging from 0 to 100. The horizontal axis labels years. | The chart reveals a dramatic increase in wildfires, especially post-1980, indicating a concerning trend. While earlier decades show minimal occurrences, there is one notable peak in 1910. | Bar | Simple Bar |
ACM_1855 copy.png | This scatter plot with error bars is titled 'Total Dataset Size = 10000 Class Distribution = 0.2, Error Rate = 0.1'. The x-axis shows Target Group Size in range of 10 to 5000, the y-axis shows FPRDifference in a range of -0.6 to 0.6. | The plot shows that the FPRDifference goes up from around -0.1 to 0 as the Target Group Size increases from 10 to 50, and the size of the error bars goes down from a 0.5 total spread to less than 0.1 after 1000 target group size. | Scatter | Scatter plot with error bars |
ACM_3081 copy.png | This multiple line chart shows the "Loss Curve" across different epochs for six models. The y-axis represents the loss, ranging from 0 to 2.5, and the x-axis represents epochs, ranging from 0 to 25000. The models include Train Loss (48, 48, 48], Validation Loss (48, 48, 48), Train Loss [64, 64, 64], Validation Loss (64... | The chart shows that the "Validation Loss" decreases as epochs increase. The Train and Validation loss 128 shows the steepest decreasepeaking at 0 epochs and dipping at 25000. The biggest loss is noted between 0 and 5000 epochs. | Line | Multiple Line |
arXiv_0767.png | This multiple line chart shows Effects of Perturbation on Preceding Layers. The x-axis shows Subsequent layers after layer 9, on a range from 0 to 20, and the y-axis shows mean 2* (AUC-0.5) after t-SNE on a scale from 0 to 1. Two conditions are compared: Bootstrap AUC, and Bootstrap Random AUC. | The chart shows that the Bootstrap AUC condition decreases from 0.7 to 1.0 on the y-axis over time. The Bootstrap Random AUC condition stays relatively stable around 1.0. | Line | Multiple Line |
pie_98.png | The chart is a pie chart titled "Most have heard 'nothing at all' about the 'alt-right'." It includes a legend with categories: "A lot," "A little," "Nothing at all," and "DK." Each category is represetend by one slice of the pie. | The majority, 54%, have heard "nothing at all" about the 'alt-right,' while 28% have heard "a little," and 17% "a lot." Only 1% are uncertain ("DK"). | Pie | Pie |
extra_0258.png | The chart is a heatmap titled "Metric: Network Clustering". The x-axis is labeled "Strategy" with three categories: Ex-ante, Ex-post, and Global. The y-axis is labeled "Algorithm" with three categories: Within-Inter Cluster, Resource Allocation Index, and Preferential Attachment. The cell represents the value of improv... | The chart shows that the Within-Inter Cluster algorithm with the Ex-ante strategy yields the highest improvement (5). The Global strategy consistently shows no improvement across all three algorithms. Other cell values have a moderate level of improvement, between 2 and 4. | Heatmap | Heatmap |
2315.png | This scatter plot with error bars shows the Team Performace over differnt tasks. THe y-axis measures the performace ranging from 0.65 to 0.8, the x-axis lists 5 task condition. | Adaptive Workflow, Direct Display, and Adaptive Recommendation have higher performance, above 075. The other two task conditions have lower performance. | Scatter | Scatter plot with error bars |
extra_0472.png | This pie chart shows the distrbution of five categories: A, C, D, Q, F. | Category A overwhelmingly dominates at 60%, while D (13%), F (12%), and Q (11%) are relatively balanced. C (4%) is notably minimal | Pie | Pie |
Springer_2nd_0888.png | This is a map titled “Rangareddy District 2021 SVM Classified Map.” It shows five land use types with colors. A compass and scale bar (0–40 km) provide orientation and distance. | The map reveals dense agricultural and forest cover in the south and center, with urban and barren areas clustered in the north and northeast. Water bodies are scattered throughout the district. | Heatmap | Heatmap |
ArXiv_3rd_790.png | This grouped bar chart is titled "F". It shows Shots Per Round on the x-axis, ranging from 0 to 60, and the probablity on the y-axis from 0.0 to 0.175. Two conditions are compared, as shown in the legend: "F" and "Human". | It shows the "F" and "HUMAN" distributions are well-aligned, both showing a similar right-skewed pattern centered around 5–10 shots per round. | Bar | Group Bar |
ACM_1980.png | This grouped bar chart shows the comparison between the AUC Gap Difference and the Test AUC Difference. The y-axis represents the Difference, ranging from -0.05 to 0.15, and the x-axis lists the Target Insititutions (Schools). Two conditions are compared for each school: AUC Gap Difference and Test AUC Difference, repr... | The chart reveals that the Difference is mostly positive across all target institutions, with the largest difference observed in School 23, where the AUC GAP Difference is over 0.15. In contrast, the Test AUC Difference is mostly minimal. | Bar | Group Bar |
arXiv_0703.png | This multiple line chart shows the success rate on the y-axis in range of 0 to 1 against the epoch on the x-axis in range of 0 to 3000. Three conditions are compared as indicated in the legend: composite, distance-based, and signal-based. | The success rate remains relatively consistent across epochs, with a slight increase in the first half of the time period. The signal-based approach achieves the highest success rate, followed by the distance-based approach, and the composite approach achieves the lowest success rate. | Line | Multiple Line |
arXiv_0035.png | This multiple line chart shows Grafting Density versus Polyethylene Glycol Molecular Weight PEG MW. The y-axis shows grafting density from 0 to 1 in chains per nanometer square, while the x-axis shows PEG MW from 0 to 40 kDa. Two types of Polyethylene Glycol are compared with four types of molecular weight, as indicate... | Grafting density decreases with PEG MW increases. AuNPs consistently show higher densities than AgNPs across all MWs. | Line | Multiple Line |
ACM_3699.png | This boxplot shows the average daily time per user on the y-axis in range of 0.1 to 100, against the number two categories on the x-axis in range . Two conditions are compared as indicated in the legend: baseline and 1s_delay. | The chart shows that the baseline condition has a higher average daily play time per user compared to the delay condition. | Boxplot | Simple Boxplot |
springer_0787.png | This stacked bar chart shows the Trend of IRSEI change in percentage for various ecogeographical regions. The x-axis lists the 11 different regions, the y-axis shows the trend of IRSEI percentage change. This change is shown using seven classes: Highly significant increase/decrease, Significant increase/decrease, Not s... | The chart shows that all regions have a trend of increasing IRSEI change. The IID2 region has the highest percentage of highly significant increase, while the OA1 region is the only region with an overall decrease in trend of IRSEI change. | Bar | Stacked Bar |
ACM_3137 copy 2.png | This boxplot shows the estimated value on the y-axis ranging from -0.5 to 2.0. The x-axis shows (c)IIPF algorithm in this spacing: a00, a01, a10, a11, π1, π2 and c. | The plot reveals that the estimated value of the (c)IIPF algorithm varies across conditions, with the highest median value at condition a00. | Boxplot | Grouped Boxplot |
1060.png | This line chart is titled 'Dischart rate vs. load'. The y-axis shows Voltage dischart rate (mV/s) in trange of 0 to 1.4, the x-axis has label Dischart current (mA) in range of 0 to 3. | The trend of voltage dischart increases with the dischart current increases. It grows faster when dischart current is increasing from 1 to 3. | Line | Simple Line |
ACM_3018.png | The x-axis represents V/(Km/h), ranging from 160 to 250, while the y-axis represents EC m/Mbps, ranging from 140 to 210. The chart includes four lines representing different devices: "P=23 IKM" (blue), "p=23 WBGRA+Stackelberg+MIMO" (black), "p=17 WBGRA+Stackelberg+MIMO" (red) and "p=17 IKM" (green), as indicated in the... | "P=23 IKM" (blue) and the "p=17 IKM" (green)" both show minimal fluctuations over V/(Km/h), although p=23 IKM is starting at 180 on the y-axis, while p= 17 IKM is only starting at 160. Both of them show a slight downward trend. p=23 WBGRA+Stackelberg+MIMO" (black) and the "p=17 WBGRA+Stackelberg+MIMO" (red) show more ... | Line | Multiple Line |
arXiv_0389.png | This scatter plot with regression line is titled State-Derivative KPI p. The x-axis shows time step in range of 0 to 150, the y-axis shows KPI in range of 0 to 0,0008. Two conditions are compared as indicated in the legend: Normal Data and Anomoly. The legend of the side of the chart represents the colour scheme, the l... | The chart shows that ‘Anomaly’ data generally exhibit higher state-derivative KPI (ρ₂₋₆) values compared to ‘Normal Data’, especially at later time steps. While the ‘Normal Data’ values remain low and stable, the ‘Anomaly’ values increase significantly over time, indicating growing deviations in system behavior. | Scatter | Scatter plot with regression line |
springer_2147.png | This line chart shows the relationship between Absorbance on the y-axis ranging from 0.0 to 1.0, and Wavelength in millimeters on the x-axis, ranging from 300 to 800 millimeters. | The line graph shows a steep increase in the 300 to 450 millimeter range, peaking around 450 millimeters at 1.0 absorbance, followed by a steep decline, falling below 0.1 absorbance. | Line | Simple Line |
ACM_2869.png | This bar chart is titled 'Proportion of Strategies'. It shows the value of proportion on the y-axis from 0% to 20%. The x-axis lists eight strategies. | The chart shows that the 'Qu' and 'Oth' strategies have the highest proportions, both exceeding 20%. While 'RP' and 'SI' have the lowest proportion, around 6%. Most strategies cluster between 7% and 17%. | Bar | Simple Bar |
Springer_3rd_051.png | This pie chart is titled 'When you interact with a chatbot or virtual assistant, do you feel that it understands your queries effectively? 80 responses.' It shows a distribution of five categories for this question: "Always', 'Most of the times', 'Sometimes', 'Rarely', and 'Never'. | The chart shows that there are 42.5% of responses were reported as 'Most of the time', 36.2% as 'Sometimes', and 10% as 'Always'. | Pie | Pie |
pie_43.png | The chart is a pie chart titled "Far more Americans favor keeping spending on policing at current levels – or increasing it – than cutting spending." It includes a legend with categories: "Increased," "Decreased," and "Stay about the same." | The largest segment, 42%, prefers maintaining current spending. A combined 31% favor increased spending, while 25% support decreased spending. | Pie | Pie |
arXiv_0833.png | This horizontal bar chart is titled "Top 20 Genres Across All Artists". The x-axis shows the number of artists, ranging from 0 to 500, while the y-axis lists various music genres. | The highest ranking genre is pop at over 500 artists, followed by electro house and dance pop at around 500. The lowest ranking is trap latino at around 300 artists. | Bar | Simple Bar |
ACM_1520.png | This is a multiple line chart chart with the axis title labeled “m subscript w d m.” The x-axis ranges from 17.5 to 35.0, and the y-axis ranges from 0.0 to 1.6. The legend includes five lines in different colours and styles. | The Adversarial ABC Posterior peaks sharply around 22.5, closely aligning with the ground-truth, shown as a narrow vertical spike. The remaining curves are broader and flatter, with peaks between 24.5 and 27.5, indicating greater uncertainty and deviation from the ground-truth. | Line | Multiple Line |
springer_0895.png | This line chart shows the relation between species rank and abundance for three classes: no propagule, moderate and intense. The x-axis shows species rank, ranging from 0 to 70, and the y-axis shows abundance, ranging from 1 to 50. | All lines start at an abundance level of 50 at species rank value 0. The chart shows a decline ins abundance as species rank increases, for all three classes. 'intense' declines the most, getting to abundance level 1 at rank 40 approximately. 'no propagule' declines the least, reaching abundance level 1 at species rank... | Line | Multiple Line |
Springer_2nd_0786.png | The chart is a line graph. It displays various methods (x-axis) against an unlabeled value (y-axis, range 0.45 to 0.85). The legend includes different markers for six strategies (SCS-1 to SCS-5) with distinct colors and line styles. | The performance metrics fluctuate across methods, with notably higher values for SCS-4 in comparison to others. Visual trends reveal a distinct peak for for method PF-ARAS illustrating higher effectiveness, while SCS-2 shows consistent lower performance, indicating less effectiveness. | Line | Multiple Line |
ArXiv_3rd_813.png | This multiple line chart is titled 'Performance'. It has performance value on the y-axis, ranging from 0 to 1, against Training Fraction on the x-axis, ranging from 0 to 1. The chart includes three lines representing different measurements, 'accuracy', 'recall', and 'precision', as indicated in the legend. | The chart shows that all three measurements have a similar trend and increase as the training fraction increases. All measurements have a little drop at 0.1 training fraction, and then keep approaching 1 afterwards. Precision has the highest performance through all training fractions. | Line | Multiple Line |
ACM_1751.png | This line chart with 4 lines shows sample size on the x-axis ranging from 0 to 1000 while the y-axis represents Average Correct Rate ranging from 0.45 to 0.95. | The line chart shows that as the sample sizes increases, the lines also increase. | Line | Multiple Line |
extra_0627.png | The chart is a box plot displaying a dollar value for various cardiac arrest interventions. It includes the legend for four categories: Ambulance AED, Pre-hospital system, CPR Bystander, and Bystander AED (incl. in-home). The x-axis is labeled "2022 USD/QALY" with a scale ranging from $0 to $500,000. | The box plots reveal variability in costs, with the Pre-hospital system showing the widest range. Comparatively, interventions like Ambulance AED exhibit lower costs, while Bystander AED (incl inhome) shows the highest median cost and some extreme outliers, as indicated by the wide whiskers. | Boxplot | Simple Boxplot |
pie_155.png | The chart is a pie chart titled "Strong Support for Army to Fight Drug Traffickers." It includes a legend with categories: Support, Oppose, and DK. The chart uses color to encode data, with percentage labels for each segment. Each category is represetend by one slice of the pie. | The chart shows 80% support for army intervention, 17% opposition, and 3% undecided. The dominant support segment highlights a strong consensus, with minimal opposition and indecision. | Pie | Pie |
springer_0443.png | This scatter plot with regression line shows the "estimated population" on the x-axis ranging from 0 to 180'000. The y-axis lists the "census population" also from 0 to 180'000. There are three lines: LanScan, WorldPop and The proposed method. | The plot shows a steep increase for all lines. The data points cluster around 40'000 to 80'000 with the lowest scores around 0 and the highest at 170'000. | Scatter | Scatter plot with regression line |
ACM_1862.png | This horizontal scatter plot is titled "T2 and Gender bias". It shows the SHAP value (impact on model output) on the x-axis in range of -0.4 to 0.2, against five different parameters on the y-axis. Each point has a corresponding Feature value as explained in the legend, ranging from Low to High. | The plot shows that C2 has the highest spread, ranging from around -0.45 to 0.3. The Feature Values are mixed for C2, while C3 and C4 have a low feature value when the SHAP value is low, and vice versa. C5 is the opposite, and C1 is always around zero SHAP value. | Scatter | Simple Scatter |
extra_0680.png | The chart is a pie chart titled "Geographical Distribution across WHO regions." The legend categorizes regions: Americas, European, Eastern Mediterranean, South-East Asian, and Western Pacific. Percentages are displayed on each segment, showing the distribution of populations. | The chart reveals that the Americas and European regions dominate with 39% and 36%, respectively, while the Eastern Mediterranean, South-East Asian, and Western Pacific regions each represent smaller shares. | Pie | Pie |
springer_4454.png | This bar chart shows the bone mineral density for three models. The x-axis shows the three models: Control, Model and ESWT. The y-axis shows the bone mineral density per unit area in percent, ranging from 0 to 100. | The chart reveals that the ESWT model has by far the highest bone mineral density, with around 75 percent. The Control and Model have close the same percentage, around 22 percent. | Bar | Simple Bar |
Springer_2nd_0512.png | The chart is a line graph titled "AAPL Apple Inc." showing the stock price of Apple Inc. (NASDAQ: USD) over the past 10 years. The x-axis represents time (years), while the y-axis displays stock price (USD) with a scale ranging from approximately 73 to 236. | The graph illustrates a significant upward trend in Apple's stock price, particularly notable after 2020, reaching a peak around 235 in 2024. This indicates strong performance, highlighting gradual price increases punctuated by fluctuations. | Line | Simple Line |
arXiv_0515.png | This is a vertical bar chart titled “XAI Current Attack Detection Feature Attribution.” The X-axis shows five features related to email and user behavior, while the Y-axis represents Attribution Score in percentage, ranging from 75 to 100 percent. Each feature is represented by a single blue bar. | Overall, features linked to sender and URL characteristics are more influential in attack detection than those related to user behavior. The feature “Email-sender anomaly” has the highest attribution score at about 91 percent. The lowest score, around 78 percent, belongs to “Login behavior anomaly.” | Bar | Simple Bar |
0726.png | This horizontal box plot summarizes the utterance sets lengths on the x-axis in range of 0 to 60, across the ten chart types on the y-axis . | Grouped Bar Chart has the highest median for utterance set length, Histogram has the lowest median for utterance set length. All boxplot has long whisker tails. | Boxplot | Simple Boxplot |
ArXiv_3rd_719.png | This multiple line is titled 'BuzzFeed 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 -10 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 -3 to -9 when the number of topics increased from 10 to 200. | Line | Multiple Line |
springer_3153.png | This multiple line chart shows the stress level (MPa) on the y-axis in range of 0 to 60 against the strain on the x-axis in range of 0 to 0.07. four lines represent different conditons as indicated in the legend. | The chart shows that stress levels increase with strain, peaking around 0.03, before gradually declining. The PA12 and PA12GB0.05 conditions show a more significant increase in stress levels, peaking around 50, while the PA12 GB0.1 and PA12 GB0.15 conditions show a smaller increase, peaking around 40. | Line | Multiple Line |
Springer_2nd_1632.png | The chart is a bar graph. It uses different colors for seasons (Winter, Pre-Monsoon, Monsoon, Post-Monsoon) to encode data points across six categories (BBS, BLS, CBL, CTC, GPL, PRD, PRI). The vertical axis represents the correlation coefficient ranging from -1 to 1. | The chart reveals varying correlation coefficients, with Pre-monsoon showing mostly high positive values, except in BLS and CBL. In contrast, the Post-Monsoon season reflects stronger negative correlations, especially in PRD. This highlights the high variance in correlation coefficients. | Bar | Group Bar |
ACM_0944.png | This bar chart plots the number of articles per year. The x-axis shows the years 2010-2024, and the y axis shows the number of articles, ranging from 0 to 10. | The number of articles published is lowest in 2010, when no articles were published. 2020 has the highest amount of articles published with 10, followed by 2023 and 2016, both with 7. | Bar | Simple Bar |
ACM_1863 copy.png | This horizontal scatter plot is titled "T3 and Text cohesion". It shows the SHAP value (impact on model output) on the x-axis in range of -0.05 to 0.05, against five different parameters on the y-axis. Each point has a corresponding Feature value as explained in the legend, ranging from Low to High. | The plot shows that C3 has the highest spread, ranging from around -0.05 to 0.08. The Feature Values are mixed for C2, while C4 has a high feature value when the SHAP value is low, and vice versa. C3 and C5 are high around zero SHAP value and low otherwise, and C1 is always around zero SHAP value. | Scatter | Simple Scatter |
pie_146.png | The chart is a pie chart titled "Would You Move to U.S.?" It includes a legend with categories: "No," "Yes, w/o authorization," "Yes, with authorization," and "DK." The chart uses percentage labels with a scale from 0% to 100%. Each category is represetend by one slice of the pie. | The majority, 61%, would not move to the U.S. Twenty percent would move without authorization, while 15% would move with authorization. A small 3% are undecided (DK). | Pie | Pie |
springer_0020 copy.png | This heatmap is titled: "(d) Vowel /a/ in not". The y-axis shows "frequency (Hz)" on a range from 500 to 10,500. "Time (s)" is shown on the x-axis, ranging from 0.05 to 0.20. There are different colors from red to blue but no caption to it. | The chart shows that the lower frequency is red. As the frequency gets higher, the color lightens up und gets from orange to yellow, green and then blue. From frequency 500 to 5500 most of the chart is between red and yellow. After that the color gets more green and just over 10,500 it turns more blue. | Heatmap | Heatmap |
springer_4159.png | This multiple line chart shows the standard deviation of the intensities in relationship to the wavelength. The y-axis shows the standard deviation of the intensities in range of 10 to the power of -1 to 10 to the power of 3. The x-axis shows the wavelength in nanometers in range of 300 to 900. Two lines are compared ... | The chart shows that the STD of measured intensities fluctuates more than STD of SNV scaled intensities and also has higher standard deviation of the intensities. | Line | Multiple Line |
ACM_3194.png | This line chart shows the error rate (%) on the y-axis, ranging from 0% to 1%, plotted against the threshold on the x-axis, ranging from 0.8 to 1.0. Two lines represent FAR and FRR, as indicated in the legend. | The chart shows a sharp decline in error rate as the threshold increases for the FRR rate, starting at an error rate of 1% at a 0.8 treshold and ending in 0% error rate at a treshold of 1.0. As the threshold increases to 0.9, the error rate drops significantly, reaching around 0.1%. The FAR condition shows the opposite... | Bar | Simple Bar |
1536c.png | This line graph, titled "Semantic Coherence," shows Semantic Coherence as a function of the number of topics (K), ranging from 3 to 20. The y-axis represents Semantic Coherence, ranging from -180 to -150. Semantic Coherence fluctuates, reaching local peaks at 3, 11, and 15 topics. | The local peaks in Semantic Coherence at 3, 11, and 15 topics indicate specific points where topic coherence is relatively high. | Line | Simple Line |
springer_0045.png | This bar chart shows "AI Processing Ability". The y-axis shows Percentage (0% to 100%) and the x-axis shows two categories: MSA Processing and Dialectal Arabic. | The chart shows that MSA processing has a higher percentage (around 64%) compared to Dialectar Arabic (60%). | Bar | Simple Bar |
springer_0101 copy 2.png | This multiple line charts is not titled. It shows the "Comparison of 3D solution results with GNSS and strong earthquake observation results (cm)" on a scale from -0.8 to 0.6 on the y-axis. The x-axis only shows numbers from 1 to 20. Two conditions are being compared: Three-dimensional deformation field by multiple InS... | The InSAR condition shows a more fluctuating trend with the highest results around 0.6 and the lowest at -0.6. The multi-source condition shows a more stable trend around 0 with a few peaks at 0.2. | Line | Multiple Line |
springer_2901 copy.png | This line chart shows the relationship between the percentage of agents with disabilities (x-axis) and the average stress level (y-axis). The x -axis ranges from 0 to 20 and the y-axis ranges from 1.8 to 2.6. | The graph shows a positive trend, with the average stress level increasing as the percentage of agents with disabilities rises. | Line | Simple Line |
Springer_3rd_023.png | This heatmap is titled 'Embedding with CNN-LSTM - Pack-per-Year Labels'. The x-axis shows predicted labels, and the y-axis shows the true labels. Both the x- and the y-axis have the same five categories: 'High', 'Low', 'Medium', 'Nonsmoker', and 'Unknown'. The cell value represents the score for classification for each... | The chart shows that the model predicts Nonsmoker (96.56%), High (88.72%), and Unknown (86.78%) accurately, with Medium (82.95%) and Low (70.86%) showing moderate confusion, mainly with each other and Unknown. Misclassifications are low overall. | Heatmap | Heatmap |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.