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_0656 copy 2.png | This boxplot is titled 'CEC2020-1' and displays the 'value', ranging from 0 to 2e10 for various algorithms. | The chart shows the highest median value for the 'COA' algorithm; 1.25. This algorithm also has the highest variance in its data, as indicated by the interquartile ranges. 'BWO' has the second highest median value; 0.75. Most algorithms show a value of 0, with no variance in the data. | Boxplot | Simple Boxplot |
arXiv_0199.png | This line chart shows two lines labeled 'infectious vector' and 'exposed vector' over time. The x-axis represents time in days, ranging from 0 to 100 days, and the y-axis represents the proportion of vector populations, ranging from 0 to 1.2e-5. | The chart shows a steep increase for both lines at first, with a peak at around 10 days. This is followed by a steady decline over the next 90 days. The line representing 'infectious vector' has a lower proportion than 'exposed vector' for all time instances. | Line | Multiple Line |
springer_2926.png | This bar chart is titled determiner use and it shows the number of errors made by students with either chinese or pakistani nationality. The x-axis represents the number of errors, ranging from 0 to 15, while the y-axis shows the number of students. | The chart shows, that only 12 chinese students made 0 errors. Most of them made at least one error. In gerenal, the pakistani students made more errors than the chinese. | Bar | Group Bar |
springer_2768 copy.png | This multiple line chart shows the shear wave (Vs) in the range of 0 to 400 on the y-axis against depth (Ft) in the range of 8254 to 8924 on the x-axis. Two conditions are compared as indicated in the legend: RF-DTSM, and DTSM_Actual. | The chart reveals that both conditions exhibit a similar trend, with a peak in the middle of the range and a sharp decline near the end. However, the "DTSM-actual" condition shows slightly higher values, suggesting greater variability and shows higher peaks. | Line | Multiple Line |
springer_1845.png | This multiple line chart shows the system average cost as a function of the computational capacity of server in GigaHertz for three different conditions: DO2MFCO, DRLOIOT, and RLIOWE. The x-axis represents the computational capacity of server, ranging from 1 to 5 gigahertz, while the y-axis represents the system averag... | The chart shows a general decline in system average cost as the computational capacity goes up across all three conditions, with the DRLOIOT condition consistently higher than the other two. | Line | Multiple Line |
ACM_1383 copy.png | This is a multiple line chart titled "Train and test loss for TIMIT." The x-axis represents the number of epochs, ranging from 0 to 50. The y-axis shows loss values ranging from 0 to 6. Two lines are shown in the legend: a blue line for "train loss" and an orange line for "test loss." | The chart shows that both training and test losses decrease steeply during the first 5 epochs, then gradually level off. The train loss continues to decline and stabilizes near zero, while the test loss plateaus at a higher value around 1.5. | Line | Multiple Line |
arXiv_0872.png | This multiple line chart shows the Avian Influenza Cases on the left side of the y-axis, ranging from 0 to 3,500,000. On the right side there is a "Post count" scale from 0 to 14,000 and a "Google Trends Score" from 4000 to 7500. The x-axis shows the date ranging from 2022-01 to 2023-03. There are five categories with ... | All categories exhibit a stagnating trend. Except for Beginning and ending of an Outbreak, where there is no line at all. The Case Numbers had a peak in 2022-04 where they reached 3,500,000. Then they decreased and stayed between 0 and 1,500,000. The X Posts had a peak in 2022-05 with 12,000 and then another one in 202... | Line | Multiple Line |
springer_0675 copy 2.png | This horizontal scatter plot shows the SHAP value (impact on model output) for various features. The datapoints which are plotted per feature are color-coded, representing low or high feature values. | The scatter plot reveals that 'Investment' has the highest impace on model output, as shown by its high SHAP value. 'inflation' has the most negative SHAP value. | Scatter | Simple Scatter |
ACM_4231.png | This is a heatmap showing pairwise differences in cosine similarity between four models: Items Only, Genre ID, Genre PE, and Random. Both axis labels are identical (Model), with values ranging from 0.0 to 0.5, encoded by a color scale and annotated within each cell. | Genre PE and Items Only show the smallest difference (0.229), suggesting similar embedding behaviors. Genre-based models diverge moderately, while the Random model consistently shows the highest differences, around 0.5, indicating it disrupts structured similarity patterns across all embedding types. | Heatmap | Heatmap |
Springer_2nd_0252.png | The chart is a stacked bar chart. The legend distinguishes between digitalization and AI solutions. The x-axis represents the years (from 2018 to 2022), while the y-axis indicates values ranging from 0 to 700. | Overall, digitalization shows a consistent upward trend over the years, while AI solutions remain significantly lower but show gradual growth. The noticeable gap highlights the dominance of digitalization compared to AI. | Bar | Stacked Bar |
springer_1698.png | This grouped bar chart shows the percentage of students in each racial/ethnic group who attempted BSTEM math from the 2017 to 2020 cohorts. The y-axis ranges from 0 to 80, capturing participation rates. | Asian students had the highest BSTEM engagement, rising from 30% in 2017 to a peak of 45% in 2019 before a slight dip to 41%. White students followed, peaking at 37% in 2019. Participation among Latino/x/a and Other groups remained moderate, between 18–20% and 24–27% respectively. Black students showed the lowest rates... | Bar | Group Bar |
ArXiv_3rd_808.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 22.99% for "actual > t" and "pred > t", 67.07% prediction for "actual < t" and "pred < t". There are a few incorrect predictions, 5.71% wrong predictions for "actual < t" and"pred > t", and 4.23% wrong predictions for "actual > t" and"pred < t". | Heatmap | Heatmap |
ArXiv_3rd_166.png | This single line chart is titled '3p2'. It shows the accuracy against the threshold. The x-axis represents the threshold values from -6 to 2, and the y-axis represents the accuracy ranging from 0.85 to 1. | The accuracy starts at around 1.00 when the threshold is -6, the accuracy decreases with many fluctuations to 0.850 until the threshold is reached 0, and then increases rapidly to 0.99 when the threshold is 2. | Line | Simple Line |
springer_2572 copy 2.png | This scatter plot shows the "UMAP_2" against the "UMAP_1". The x-axis represents UMAP_1 values ranging from -10 to 10, and the y-axis represents UMAP_2 values ranging from -4 to 4. Different conditions are compared in the graph. | The chart shows that the CD8+T cells have the highest UMAP amount, while Neutrophils have the least amount of UMAP. | Scatter | Simple Scatter |
ArXiv_3rd_182.png | This single line chart is titled '2p3'. It shows the accuracy against the threshold. The x-axis represents the threshold values from -6 to 2, and the y-axis represents the accuracy ranging from 0.92 to 1. | The accuracy starts at around 0.98 when the threshold is -6, the accuracy decreases with lots of fluctuations to 0.92 until the threshold is reached 0, and then increases rapidly to 1 when the threshold is 2. | Line | Simple Line |
ArXiv_3rd_3.png | This multiple line chart shows Evaluation Metrics - Graph. The x-axis lists 6 method categories, and the y-axis shows evaluation values ranging from 0.70 to 0.95. 4 conditions are compared: Accuracy, Precision, Recall and F1-score. as indicated in the legend. | The chart shows that precision and accuracy metrics are generally higher for all methods than recall and F1-score metrics. SVM method shows the best performance on all four metrics conditions, all above 0.85. CNN shows the worst performance on all four metrics, all below 0.85. | Line | Multiple Line |
1295.png | This is a boxplot chart displaying mean values for pre and post measurements of three scales: PANAS POS (positive emotions), PANAS NEG (negative emotions), and GSE (general self-efficacy). The y-axis represents scores ranging from 0 to 50. | The difference between pre and post measurements for PANAS POS is significant at a level of p<0.01, while for GSE it is significant at a level of p<0.05. | Boxplot | Grouped Boxplot |
springer_1105 copy 2.png | This scatter plot is titled 'all women' and shows the risk ratio estimate for two metrics: 'francophone' and 'angolphone'. Per metric, three time intervals are shown: 2008-2011, 2012-2015 and 2016-2020. The y-axis shows the time intervals, the x-axis shows the risk ratio (95% CI) estimate. | The chart shows a higher risk ration estimate for later time intervals: for francophone, the estimates are: 1 for 2008-2011, 1.32 for 2012-2025 and 1.70 for 2016-2020. For anglophone, these estimates are: 1.07 for 2008-2011, 1.40 for 2012-2025 and 1.82 for 2016-2020. | Scatter | Simple Scatter |
extra2_0159.png | The chart is a pie chart titled "Do you think gynecologic robotic surgery for malignant conditions benefits patients?" It encodes responses as proportions, with a legend indicating five response categories: "Strongly agree," "Somewhat agree," "Somewhat disagree," "Strongly disagree" and "Neutral/ Not sure". The axis la... | A majority (75.4%) of respondents express agreement with the benefits of gynecologic robotic surgery, indicating a strong positive perception. In contrast, only 21.6% of respondents disagree, revealing a significant trend favoring the procedure's effectiveness for malignant conditions. The data illustrates a clear pref... | Pie | Pie |
Springer_3rd_131.png | This heatmap is titled 'Correlation Matrix'. Both the x-axis and the y-axis have the same seven categories: 'pH', 'N', 'P', 'K', etc. The cell value represents the score of similarity, ranging from 0.0 to 1.0. | The plot reveals that the 'pH' variable is strongly negatively correlated with 'N', 'P', and 'K' (-0.71), while N, P, K, and OC are highly positively correlated (0.80–0.86), forming a tight cluster. There are moderate negative correlations with N, P, K, and OC (-0.48 to -0.67) but a weak positive correlation with pH (0... | Heatmap | Heatmap |
pie_31.png | The chart is a pie chart titled "Colombia: share of historical site visitors 2017, by age group." It includes a legend with age groups: 12 to 25, 26 to 40, 41 to 64, and 65+. Each age group is represetned by one slice of the pie. | The 26 to 40 age group has the largest share at 31.2%, followed closely by 12 to 25 at 30.2%. The 41 to 64 group accounts for 23.5%, while 65+ is the smallest at 15.1%. | Pie | Pie |
ACM_2978 copy 3.png | This bar chart shows the synergy value of channels. The x-axis lists channels from 1 to 12, and the y-axis shows synergy value ranging from -0.5 to 1. | The chart shows that the first channel has no synergy value, while the channel 3 and 4 have the highest energy value, around 1. The second channel has a slightly higher energy value than the first channel. | Bar | Group Bar |
springer_1614.png | This pie chart shows the distribution of mobility destinations. Internal migration is the most common, accounting for 51% of cases. Movements that span both internal and international borders follow at 21%, while international-only mobility represents 20%. A small segment, 8%, is labeled as unspecified. | The predominance of internal migration underscores its significance in mobility patterns, while the relatively balanced proportions of international and dual-destination moves suggest diverse migration dynamics. The unspecified portion, though minor, points to a gap in data precision. | Pie | Pie |
ArXiv_3rd_581.png | This multiple line chart shows the number of RIS Reflecting Elements on the x-axis, ranging from 10 to 100. The throughput is shown on the y-axis, ranging from 0 Kbps to 100 Kbps. Six conditions are compared and indicated in the legend. | The chart shows that BC-WPUSNs and WPUSNs remain nearly flat in throughput with the increase in the number of RIS reflecting elements. Other conditions have an increasing trend in the throughput. RIS-aided WPBUSNs (Optimal) achieving the highest throughputs, surpassing 1100 Kbps at 100 RIS reflecting elements. | Line | Multiple Line |
springer_0600 copy.png | This bar chart shows 'support services' across three conditions: shallow, intermediate, and deep. The y-axis represents support services, ranging from 0 to 0.3. Each bar shows error bars as well. | The shallow group shows the highest support service of 0.19, followed by intermediate with 0.14 and deep with a support service of 0.1. | Bar | Simple Bar |
extra_0578.png | The chart is a scatter plot plotting EHI vs. RSEI, showing the relationship between the EHI (y-axis) and RSEI (x-axis). The legend indicates the density of points using a gradient. The axes are labeled "RSEI" (0 to 1) and "EHI" (0 to 1), with a black exponential trend line illustrating the correlation of 0.72. | The scatter plot reveals a strong positive correlation between EHI and RSEI, indicated by a trend line that follows an exponential pattern. The fitted curve is "y = 0.3713 * exp(0.8361 * x)". | Scatter | Simple Scatter |
springer_1696.png | This dual-axis line chart compares the SCI300 Index (left y-axis, dark gray line ranging from 0 to 4000) and Investor Sentiment (right y-axis, light gray line) over a timeline centered on a vertical dashed event marker at time 0. The x-axis ranges from -24 to +24 time units, possibly months. | Both series show a notable upward spike around the event before diverging: the SCI300 remains relatively stable post-event, fluctuating around 3200–3400, while Investor Sentiment declines steadily from a peak near 70 to a low around 40. This suggests the event triggered an initial surge in market optimism that wasn’t s... | Line | Multiple Line |
springer_3231.png | This is a scatter plot showing Peak Ground Acceleration (PGA in m/s²) from 0.01 to 10, versus Hypocentral Distance (km) from 1 to 1000 km. Points are color-coded by earthquake magnitude ranges: gray (4‒5), blue (5‒6), green (6‒7), and red (7‒8), as indicated in the legend. | The chart shows that PGA generally decreases with increasing distance. Higher magnitude events (red and green) maintain higher PGA at longer distances, while lower magnitudes (gray and blue) cluster at lower PGA and closer ranges. | Scatter | Simple Scatter |
arXiv_0547.png | This line graph titled “Homomorphic Encryption Security Analysis” plots key length (bits) on the x-axis (1000 to 8000), with two y-axes: time to breach (log scale, blue, from 10⁶ to 10¹⁴ hours) and success rate of attacks (%, red, from 10 to 70%). | As key length increases, the time to breach rises exponentially while the success rate of attacks sharply declines, indicating stronger security with larger key sizes in homomorphic encryption. | Line | Multiple Line |
springer_2557 copy 11.png | This bar chart shows the Frequency on the y-axis, ranging from 0 to 0.4, against the CN on the x-axis, ranging from 75 to 100. | The chart shows a concentrated distribution between 85 and 90 TWI with a Frequency peak at over 0.4. | Bar | Simple Bar |
Springer_2nd_1735.png | This bar chart presents Root N:P ratios across four groups (G, SH, ST, F), with the vertical axis ranging from 0 to 50. | Groups SH and ST show the highest and statistically similar values at around 22–23, marked “a”. Groups G and F display lower ratios, both labeled “b”, with values near 17 and 15 respectively. The distribution suggests significant differences in nitrogen-to-phosphorus balance between treatment groups, with SH and ST hav... | Bar | Simple Bar |
ACM_1878.png | This multiple line chart shows the run time in seconds on the y-axis in range of 0 to 160 against the class size on the x-axis in range of 0 to 1000. The chart includes five lines representing different conditions: DRR, GRR, Priority, Random, and Group Matcher. | The chart shows that the Priority line has slower run times, rising linearly from almost zero to around 160 seconds from class size 20 to 1000. All other conditions have faster run times, with only GRR going up to 10 seconds and the rest staying below 5. | Line | Multiple Line |
ArXiv_3rd_493.png | This multiple line chart is titled 'Behavior Content on Twitter. It shows the F1 score on the y-axis in the range of 10 to 60, against the History Size on the x-axis in the range of 10 to 70. Three conditions are compared as indicated in the legend: Zero-shot, Std-CoT-FT, and OM-CoT-FT. | The 'OM-CoT-FT' method achieves the highest and most stable performance, peaking at history size 30. The 'Std-CoT-FT' method closely follows a similar pattern but shows more fluctuation. The 'Zero-Shot' method performs the lowest across all the history sizes. | Line | Multiple Line |
springer_3161.png | This scatter plot with regression line shows Residuals on the y-axis in range of -300 to 300, against Predicted values on the x-axis in range of 0 to 550. | The scatter plot with regression line suggests a positive relationship between the residuals and predicted values. With the scatters arranging in the middle of the chart across the regression line. | Scatter | Scatter plot with regression line |
springer_2482 copy.png | This bar chart shows the TC (mmol/L) on the y-axis, ranging from 0 to 4, against the three different models: "CK", "ZEN", and "ZEN+ZJ-2019-1". | The chart shows the "CK" and "ZEN+ZJ-2019-1" model are at around 2.9 TC while the "ZEN" is at about 3.3. | Bar | Simple Bar |
Springer_2nd_0222.png | The multiple line chart three lines representing different times: "15 minutes", "30 minutes", and "45 minutes". The x-axis is labeled "percent of Spent Garnet addition (G lower add) [percent]" ranging from 0 to 100 and the y-axis is labeled "sigma lower max [kPa]" ranging from 0 to 20. | The multiple line chart reveals a decreasing trend for all three categories, suggesting a negative correlation. | Line | Multiple Line |
arXiv_0126.png | This boxplot is titled 'Box Plot with Individual Points'. The y-axis shows F1 score in the range of 0.6 to 0.9, the x-axis shows two conditions: XGBoost, and ADAM. | The F1 score for XGBoost has a higher median and wider range than that of ADAM. XGBoost has a median F1 score of around 0.75, while ADAM has a median F1 score around 0.65. | Boxplot | Simple Boxplot |
pie_60.png | The chart is a pie chart titled "Most Americans say they’ve heard nothing at all about QAnon." It uses color to encode responses: "A lot," "A little," "Nothing at all," and "No answer." Each category of responses is represetend by one slice of the pie. | The majority, 76%, have heard nothing about QAnon, while 20% have heard a little, and 3% a lot. A small 2% did not answer, highlighting a significant lack of awareness. | Pie | Pie |
Springer_2nd_1830.png | This is a line graph showing "Coherence score (UMass)" ranging from 0.04 to 0.54 on the y-axis and "Number of topics" ranging from 10 to 20 on the x-axis. A vertical red dotted line marks topic number 13 with a label. | The graph shows that the coherence score peaks at 13 topics, suggesting this is the optimal number for topic modeling. Scores fluctuate afterward, indicating reduced topic coherence beyond this point. | Line | Simple Line |
springer_0373.png | This multiple line chart shows the relationship between "Porosity" on the x-axis, ranging from 0 to 0.25 and "V (m/s)" on the y-axis ranging from 3000 to 6500. There are three conditions being compared: Wang et al. 2015, SCA and DEM. There are four different lines and colors. | The chart shows a general decreasing trend in V, as the Porosity gets higher. The highest V has AR=1 and the lowest has AR=0.08 | Line | Multiple Line |
ArXiv_3rd_187.png | This scatter plot with regression line is titled "2p4", 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 aligned with the line 'y=x'. | Scatter | Scatter plot with regression line |
extra_0711.png | The chart is a box plot titled "Movement Competence." It features two groups: Intervention and Control, with two time points (Baseline and 12 weeks). The Y-axis is labeled "Movement Competence" and ranges from 0 to 100. | The box plot reveals that the intervention group shows a lower decrease in movement competence from baseline to 12 weeks, while the decrease in the control group is much larger. This suggests that intervention might lead to a slower reduction of movement comptence. | boxplot | Grouped Boxplot |
arXiv_0131.png | This bar chart shows Hospitalization in Past 6 Months. The y-axis shows count in range of 0 to 400, the x-axis shows Hospitalization in the range of 0.0 to 6.0. | The chart shows that the majority of individuals (390 people) have not been hospitalized in the past 6 months. Among those who have, most (around 100 people) have only been hospitalized once. Very a few people have been hospitalized more than twice. | Bar | Simple Bar |
arXiv_0821.png | This multiple line chart shows the "average discounted return" on the y-axis in range of 0 to 8. The x-axis is titled "k" and shows numbers in range of 0 to 150. There are seven different conditions: Centralised, Independent, Networked (0.2), Networked (0.4), Networked (0.6), and Networked (0.8) and Networked (1.0). | All "Networked" conditions show a significant increase in the beginning , with a following stable trend. They range between 7 and 6 with Networked (0.4) having the highest average discounted return and Networked (1.0) the lowest. The Independent category stays around 1.5 and the Centralised category shows a morge fluct... | Line | Multiple Line |
arXiv_0286.png | This line chart shows the "HybriDNA Training Loss" as a function of "Training Tokens" on the x-axis, ranging from 0 to 250 billion tickets, and "Loss" on the y-axis, ranging from 1.025 to 1.200. Two lines represent HybriDNA 300M (Hybrid) in blue and HybriDNA 300M (pure Mamba) in orange, as indicated in the legend. | The chart shows a clear divergence between the two conditions, with the "HybriDNA 300M (pure Mamba)" condition consistently outperforming the "HybridDNA 300M (Hybrid)" condition across all training levels. The initial loss for HybridDNA 300M (pure mamba) is notably higher, peaking around 1.200 before steadily declining... | Line | Multiple Line |
1753b.png | The horizontal bar chart titled "Expenditure" compares the financial activities of Managers and Scholars across six categories: Breeding new Axies, Convert into PHP, Convert into stablecoins, Other, Purchasing goods and services in the real world, and Trading into other cryptocurrencies. The x-axis represents the count... | Convert into PHP is the most reported expenditure category for both Managers and Scholars. Same amount people reports Purchasing goods and services in the real world (SLP/AXS payment). | Bar | Group Bar |
springer_1139.png | This multiple line chart plots 'Temp cel' on the x-axis, ranging from 40 to 540, against DTG (mg/min) on the y-axis, ranging from 0 to 100. Six lines are plotted in the graph, labeled either WPS or HEPS. | The chart shows similar behavior for all curves. There is a stable level around 0-10 DTG. At temp cel value approximately 300 all the lines spike up to 100% and drop again to the stable level afterwards. | Line | Multiple Line |
springer_1704 copy 2.png | This dual-line chart displays adsorption and desorption behavior as functions of relative pressure. The x-axis represents relative pressure (p/p⁰), ranging from 0 to 1, while the y-axis shows the quantity adsorbed in cm³/g STP, extending beyond 600. | Both curves: adsorption (blue) and desorption (orange), rise gradually up to a pressure of about 0.8, after which they sharply spike. The loop between the two lines suggests hysteresis, a typical feature in mesoporous materials, indicating that adsorption and desorption processes are not fully reversible. | Line | Multiple Line |
Springer_2nd_0352.png | The chart is a bar chart. The x-axis is labbeled "Decrease in PANAS NA score" and ranges from 0 to 6, while the y-axis compares two groups: TippyMirror-O and TippyVR-Plus. | The chart suggests that TippyMirror-O shows greater decrease in PANAS NA scores, indicated by a longer bar compared to the bar for TippyVR-Plus. | Bar | Simple Bar |
ACM_1591 copy 3.png | This is a line chart titled “OrganSMNIST.” The X-axis represents four blocks labeled “Block 1” through “Block 4,” and the Y-axis shows accuracy ranging from 0.79 to 0.93. In the legend, four lines, each with a distinct color and marker, represent image resolutions : 28 by 28, 64 by 64, 128 by 128, and 224 by 224 pixel... | All image resolutions show increasing accuracy across blocks. The 64 by 64 resolution achieves the highest accuracy in the final block. Although 224 by 224 starts with the lowest accuracy, it shows the steepest improvement across blocks. | Line | Multiple Line |
ACM_0269.png | This horizontal stacked bar chart is not titled. The x-axis represents the percentage of students with 0% to 100% and the y-axis lists four categories: Mid (Fa23), End (Fa23), Mid (Sp24) and End (Sp24). Each bar is segmented into three levels of agreement: Strongly disagree, Neither agree nor disagree and Strongly agre... | The chart shows an increasing trend of students strongly agreeing from Mid (Fa23) to End (Sp24). In the categories Mid (Fa23), End (Fa23) and Mid (Sp24) most of the students neither agree nor disagree. In the category End (Sp24) the biggest segment of the bar is strongly agree (42%). | Bar | Stacked Bar |
Springer_3rd_077.png | This pie chart shows the distribution of five categories: 'Very Low', 'Low', 'Moderate', 'High', and 'Very High'. | The chart shows that the category of 'Very High' is dominant with 68%, and the 'High' and 'Moderate' category has 14% each. Other categories have a lower percentage. | Pie | Pie |
1501.png | This grouped bar chart displays error counts on the y-axis, ranging from 0 to 120, for three input techniques (Dwell time, Pursuits, and Gaze gestures) across sitting and walking states. The x-axis is categorized by these techniques, with each technique further divided into sitting and walking states. The legend distin... | Across all three input techniques, errors are consistently lower in the sitting state. The chart shows that Gaze gestures are less error-prone and remain unaffected by the number of targets. In contrast, errors increase with a greater number of targets when using Dwell time and Pursuits. | Bar | Group Bar |
extra_0556.png | The chart is a pie chart. The legend indicates categories such as Forest, Grass land, Farmland, Land for construction, Waters, and Unused land. The chart encodes percentages with segments representing various land types. | The chart reveals that Forest comprises the largest share at 41.48%, while Unused land shows a negligible proportion at 0.002%. Grass land (28.78%) and Farmland (24.50%) are the other two major segments, highlighting the dominance of forest, grassland and farmland. | Pie | Pie |
arXiv_0184.png | This heatmap is titled 'Consistency map (Downsampled) mVAWG'. It shows time in seconds on the x axis, ranging from 250 to 2000 seconds. Repeat number is shown on the y-axis, ranging from 0 to 50. The heatmap shows voltage for each combination of time and repeat. | The map shows a voltage close to 0 for all values, except right after 500 seconds. There, a spike to 0.25 volts is shown, followed by a decrease immediately afterwards that gradually returns to the basis level. This happens for all repeat numbers, so this change is shown as a vertical line. | Heatmap | Heatmap |
springer_1337 copy 2.png | This bar chart presents different binding energy components (x-axis) with corresponding energy values on the y-axis, ranging from –150 to +150 kcal/mol. | AG Bind has the lowest energy, around -100 Kcal. AG Bind SolvGB has the highest energy, around -20 Kcal. | Bar | Simple Bar |
ACM_1779 copy 2.png | This scatter plot shows the relationship between the F1-Score on the x-axis ranging from 0.78 to 0.92 and the AUCROC Score on the y-axis ranging from 0.875 to 0.975. The plot includes four different classifiers: KNN, XGBoost, Random Forest and Decision Tree. each classifier has five datapoints, using five different Gen... | The distributions of the data suggests that Random Forest has the highest F1-Scores of above 0.9 for all generators except CTGAN, which is around 0.84. All datapoints have an AUCROC Score over 0.925 except KNN using CTGAN, and Decision Tree using Real Data, CTGAN and DistilGPT. The worst performance is that of the Deci... | Scatter | Simple Scatter |
springer_2842 copy.png | This scatter plot with regression line is titled '(e) Paracatu river basin', it shows the anomaly (mm year) on the y-axis in range of -600 to 800, against the years on the x-axis in range of 1985 to 2020. | The chart shows a general increase in the frequency and magnitude of negative precipitation anomalies over time in the Paracatu river basin, with notable positive peaks in years such as 1983, 1990, and 1993. | Bar | Simple Bar |
springer_3756.png | This scatter plot with a fitted linear regression line examines the relationship between cumulative fiscal consolidation (2011–2013, % of GDP) on the x-axis ranging from 0 to 20, and cumulative real GDP growth (2011–2013, %) on the y-axis ranging from -20 to 10. The equation of the regression line is y = -1.11x + 5.40... | A negative linear trend indicates that higher fiscal consolidation correlates with lower GDP growth. Greece, has the highest consolidation, experienced the lowest GDP growth, around -15%. Luxembourg, has minimal consolidation with the highest GDP growth, around 5%. | Scatter | Scatter plot with regression line |
ACM_0186.png | This multiple line chart is titled "SQL" Advanced. The x-axis shows numbers ranging from 0 to 6, while the y-axis shows numbers ranging from 0 to 1. The chart includes two lines representing "Present" and "Not present" conditions, as indicated in the legend. | The chart shows a positive trend, with both "Present" and "Not present" increasing. The "Not Present" line consistently stays above 0.5, while the "Not present" line remains below 0.7 until they reach number 4 on the x-axis. After that both lines increase to 1 on the y-axis. | Line | Multiple Line |
1781c.png | This line chart shows Effects of EMS on Force Application (Target Force = 9N). The y-axis shows force(N) in the range of 0 to 10, the x-axis shows time (s) in range of 0 to 10. Three conditions are compared: With EMS, Without EMS, and Target. Each condition is reprented by a line. | The target condition is stablized at 9N with time increases. With EMS condition increases force from 0N to 9N at the beginig, and then stayed around 9N afterwards. Without EMS condition increases force from 0N to 13N at the begining, and then fluxtuated around 13N afterwards. | Line | Multiple Line |
springer_3042.png | This heatmap is titled ‘Seasonal Correlation of Hydro-ecological Parameters’. Both the x-axis and y-axis display six hydro-ecological parameters, including water coverage, NDTI, and NDVI during pre- and post-monsoon seasons. Each cell is color-coded to represent the correlation coefficient between pairs of parameters, ... | The chart indicates that NDTI (post-monsoon) has the highest positive correlation with water coverage (pre-monsoon) at 0.65, followed by NDTI (pre-monsoon) with water coverage (post-monsoon) at 0.49. NDVI (pre-monsoon), on the other hand, shows a strong negative correlation with NDTI (post-monsoon) at -0.77 and NDVI (p... | Heatmap | Heatmap |
springer_4369.png | This line chart shows the relationship between the true positive rate and the false positive rate. The x-axis represents the false positive rate, ranging from 0 to 1. The y-axis represents the true positive rate, ranging from 0 to 1. The line is called 2019 Cohort, as seen in the legend. The area under ROC curve is 0.8... | The chart shows a positive trend, the true positive rate increases as the false positive rate increases. The first increase is the biggest. | Line | Simple Line |
Springer_2nd_1738.png | The chart is a scatter plot showing Health Workers vs. Age-standardized Prevalence Rate for Dental Personnel (2019). The x-axis represents the number of health workers per 10,000 population, while the y-axis shows the age-standardized prevalence rate per 100,000 population. The legend includes data points for various c... | The scatter plot reveals a negative correlation (-0.41) between the number of dental health workers and the prevalence rate, suggesting that countries with fewer dental personnel often report higher prevalence rates. | Scatter | Scatter plot with regression line |
Springer_3rd_035.png | This pie chart shows the distribution of three categories: 'Recall', 'Precision', and 'Accuracy'. Each category is represented by one slice of the pie, with the percentage labelled. | The chart shows that Accuracy has the highest percentage, at 34.5%. The Precision category follows at 33.8%. The Recall category has 31.7%. | Pie | Pie |
1596.png | This set of bar charts displays four performance metrics: Accuracy, F1 Score, Precision, and Recall on the y-axis, with separate bars representing text concepts (blue) and image concepts (orange) for each metric. The y-axis values range from 0 to 1. Error bars are shown for each bar to indicate variability in the resul... | Across the charts, image concepts generally perform better in terms of accuracy, while text and image concepts show more comparable results in F1 score, precision, and recall, with some overlap in error bars indicating potential variability in the performance differences. | Bar | Group Bar |
springer_2518 copy.png | This multiple line chart is titled "Occipital Lobe" and shows the relationship between the "K" on the x-axis, ranging from 2 to 8, and the "MAE" on the y-axis, ranging from 3.5 to 5.5. The chart includes four lines representing different models: "PQR", "PLS", "APQR", and "QRfPC". | The plot shows that nearly all models exhibit a positive trend, with the MAE increasing as K increases. However, the "QRfPC" and "PQR" condition remains stable. | Line | Multiple Line |
ACM_1926.png | This multiple line chart titled "ROC-AUC" shows the "True Positive Rate" on the y-axis, ranging from 0 to 1, plotted against the "False Positive Rate" on the x-axis, ranging from 0 to 1. The chart compares two conditions: "ROC-AUC = 0.463 (BKT)" and "ROC-AUC = 0.811 (BERT)". | The chart shows that the BERT true positive rate increases gradually with the false positive rate, being consistently concave down, making the area under the curve substantially higher than 0.5. The BKT true positive rate increases gradually at the start, then stays constant at around 0.43 for most false positive rates... | Line | Multiple Line |
springer_0419.png | This multiple line chart shows the relationship between "E/V" on the x-axis, ranging from 0.1 to 0.5 and "I/A" on the y-axis, on a scale from 0 to 1.6. There are two different conditions: "SPCE/AgNPs" and "SPCE". | The SPCE/AgNPs condition shows a steep increase with a peak at 1.5 and a steep decrease to 0 again. The SPCE condition shows a smaller increase later in E/V. | Line | Multiple Line |
arXiv_1220.png | This multiple line chart shows the 95% energy containment of (Ekin - E)/ Ekin, with the x-axis representing the Kinetic energy (GeV) ranging from 0 power of 2 to the power of 4, and the y-axis representing the Energy containment ranging from 0.0 to 0.050. The chart includes four lines representing Non-fiducial Erec, No... | The non-fiducial Erec has the highest energy containment, with an increasing trend from 0.040 to 0.050 as the kinetic energy increases. The other three conditions show lower and stabilized energy containment values as the Kinetic energy increases. | Line | Multiple Line |
arXiv_0855.png | This pie chart is titled "Duration (s)". The chart uses color to encode responses, with darker shades representing higher scores. Each category is represented by one slice of the pie and marked with the percentage they take up. | The chart shows that (6,8) is the most common duration, with 47.4%, followed by (2,4) at 32.8% and (4,6) is less common, with 19.8%. | Pie | Pie |
springer_2190.png | This scatter plot shows the estimated LE8 risk on the y-axis in range of -6 to 4 against metals in micrograms per Liter on the x-axis ranging from 0.00 to 1. | The plot shows a clear negative trend between estimated LE8 risk and the amount of metal. As the number of metals increases, the estimated LE8 risk decreases, indicating a strong inverse relationship. | Scatter | Simple Scatter |
Springer_2nd_0802.png | The chart is a horizontal scatter plot showing "Author Contributions Over Years." it features a line between two points or every author, representing a time interval. The x-axis is labeled "Year" (2001 to 2023), while the y-axis lists authors. Both axes use a linear scale. | The scatter plot shows varying contributions from authors over time, highlighting trends with larger points indicating significant contributions. Notably, there is increased activity in the years 2019-2021, as shown by the bars. Smaller points reflect lesser contributions, indicating variability in authors' engagements... | Scatter | Simple Scatter |
ACM_3688.png | This bar chart shows the movement time of two conditions: Non-PD and PD. The y-axis represents movement time (ms), ranging from 0 to 1400 ms, and the x-axis shows the two conditions. | The chart shows that the "PD" group has a median movement time of 1325.67ms, while the "Non-PD" group has a median time of 786.62ms. Additionally, the interquartile range for the "PD" group is wider than that of the "Non-PD" group. | Bar | Simple Bar |
pie_141.png | The chart is a pie chart titled "A Majority of Adults Have Received Benefits." It includes a legend with two categories: "Ever received benefits" and "Never received." Each category is represetend by one slice of the pie. | The chart shows that 55% of adults have received benefits, while 45% have not. This indicates a majority of adults have experienced receiving benefits, highlighting a significant societal trend. | Pie | Pie |
Springer_2nd_1670.png | This line chart shows one-year survival probabilities by era and preservation method. The x-axis ranges from 0 to 360 days, and the y-axis shows survival probability from 0.7 to 1.0. Four groups are compared: Early–Prolonged, Early–Standard, Modern–Prolonged, and Modern–Standard. | The Modern–Standard group consistently shows the highest survival, staying near 0.9, while Early–Prolonged drops to around 0.75. The log-rank test shows p < 0.0001, confirming statistically significant differences. Survival is highest with modern techniques and standard preservation times. | Line | Multiple Line |
arXiv_0545.png | This is a vertical bar chart. The x-axis shows three resource type metrics, and the y-axis shows usage percentage ranging from 0 to 40%. Two color-coded bars represent each category, as indicated in the legend. | The chart shows that the PET Framework consumes significantly more CPU and Battery but less Memory than Standard Operations. CPU usage rises from 12% to 31%, Memory drops from 39% to 24%, and Battery usage increases by 8%. | Bar | Group Bar |
ACM_2353.png | This multiple line chart shows the correlation of GT and Exracted. All charts show a range of 0-160 on the x-axis and 0.0-1.0 on the y-axis. The charts include two lines representing the Gt in orange and Extracted in blue. | The chart reveals that both GT and Extracted processes exhibit a general similarity in patterns. | Line | Multiple Line |
Springer_2nd_0169.png | The chart is a bar graph showing Subjective Confidence vs. Subjective Loss Probability. The x-axis represents "Subjective Loss Probability," ranging from 0.6 to 0.8, while the y-axis indicates "Subjective Confidence," scaled from 0% to 100%. | The chart illustrates a pronounced peak in subjective confidence (90%) at a loss probability of 0.7, indicating strong consensus among participants. Outside this range, confidence sharply declines to 0%, indicating a strong concentration of the confidence at probability level 0.7. | Bar | Simple Bar |
0949a.png | This horizontal scatter plot with error bars is titled "MCS: Text c Chart Accuracy Means", 4 categories is listed on the y-axis, Text: Paragraph, Text: list, Chart: Cartisian, Chart: Polar. the x-axis is in range 0.00 to 1.00. | All 4 categories have accuracy means in the range of 0.75 to 1.00, Chart: Cartesian category has the highest accuracy means. Text: paragraph category has longest error bars, while Chart: Polar category has shortest error bar. | Scatter | Scatter plot with error bars |
extra_0227.png | This grouped boxplot shows the value of Zooplankton / Phytoplankton on the y-axis from 0 to 28, against the time (days) on the x-axis from 0 to 56. Three conditions are compared, as indicated in the legend: Initial, Control, and S.woodiana. | The chart shows that Initial ratios are high but variable. From day 14. Both conditions have a drop in the value of Zooplankton / Phytoplankton, the control conditions show larger drop than the S.woodiana condition. | Boxplot | Grouped Boxplot |
springer_0568.png | This bar chart shows the % of total fatty acids on the y-axis, ranging from 0 to 25. The x-axis lists various categories. | The highest total amount of fatty acids is shown in the C:16:0 category with around 23%. The lowest is in the C14:0 category with 1%. | Bar | Simple Bar |
Springer_2nd_0382.png | The chart is a multiple line graph titled "Literacy rate: Global adult literacy rate aged 15 years and older from 1976 to 2022 by gender." It features three lines representing overall, male, and female literacy rates. The x-axis is labeled with years from 1976 to 2022, while the y-axis shows literacy rates from 0 to 10... | Overall literacy rates have steadily increased since 1976, reaching nearly 90% by 2022. Male literacy rates remain slightly higher than female rates, though both show similar upward trends, indicating ongoing improvements in global education access for both genders. | Line | Multiple Line |
extra_0322.png | This multiple line chart shows the value of Annual ME Production (million tons) on the y-axis from 30 to 380, against the production year on the x-axis from 1960 to 2030. Two lines are compared: Linear trendline and Annual production growth line. | The chart shows that the annual ME production has a consistent long-term growth with fluctuations around the linear trendline. Production lagged behind the trend before 2010 but accelerated afterward, surpassing the expected linear increase. | Line | Multiple Line |
springer_0161.png | This multiple line chart is titled Olivine S. The y-axis shows Pr (%) from -10 to 30 and the x-axis shows the phase angle a (°) on a range from 0 to 180. Two conditions are being compared: 442nm and 633nm | Both lines start at -5 Pr%, then increase to around 10Pr% at phase angle a 80° and then decrease to 0 again. The 633nm condition (peak at 10) slightly outperforms the 442nm condition with peak at 8. | Line | Multiple Line |
0961.png | This grouped bar chart is titled "PANAS-X Negative Affect Outcome Measurement". The y-axis shows Measurement Score in the range of 1 to 5, the x-axis shows 4 conditions represented by 4 bars. | In the Black group, low condition has 1.25 score compared to high condtion with 1.75 score. In the white group, the low condition has score1.19, and high condition has 1.45 score. This indicates that Black groups have higher PANAS-X Negative affect compareed to the white group. | Bar | Group Bar |
extra_0371.png | This heatmap shows the predicted label on the x-axis and the true label on the y-axis. Both the x- and y-axes have the same categories from 0 to 6. The cell value represents the counts of predictions for each label pair, and a color bar is listed on the right side indicating the range of predictions is from 0 to 30000. | The chart shows high prediction accuracy. The class '0' has the highest correct prediction, of 34533 count, and the class '4' has the lowest correct prediction, of 2556 count. | Heatmap | Heatmap |
Springer_3rd_130.png | This heatmap is titled 'Heatmap for Calculating Similarity using BERT, Only full texts'. Both the x-axis and the y-axis have the same six categories: 'Anthropic', ChatGPT', 'Gemini', 'Llama3', 'Mistral', and 'Perplexity'. The cell value represents the score of similarity, ranging from 0.0 to 1.0. | This plot reveals that the models exhibit uniformly high similarity (0.90–0.98) on full texts, with ChatGPT and Perplexity, and Gemini and Perplexity reaching the highest (0.98). Llama3 shows slightly lower similarities (0.90–0.94). | Heatmap | Heatmap |
ArXiv_3rd_39.png | This bar chart is titled 'Ranked general population responses on how attributes might impact AI behavior for Africans'. The x-axis lists 20 attributes, and the y-axis indicates the percentage of respondents, ranging from 0% to 35%. | The graph shows that the category 'Colonial history of country of origin' has the highest percentage of respondents, around 32%, and the category 'Age' has the lowest percentage of respondents, only 15%. | Bar | Simple Bar |
springer_1064.png | This scatter plot shows principal component 1 and 2 on the x- and y-axis, ranging from -15 to 5 and -10 to 10 respectively. Various yeast species represent datapoints in the plot. | The plot reveals that 'P. pijperi' species has the highest principal component 2, between 5 and 8. Most yeast species have a principle component 1 greater than 0. | Scatter | Simple Scatter |
Springer_2nd_1629.png | This is a line graph . The x-axis shows epochs from 0 to 100, and the y-axis shows loss values from 0.7 to 1.0. The legend differentiates a solid purple line for training loss and a dashed yellow line for validation loss. | The graph shows that training and validation losses decrease rapidly in the initial epochs and then stabilize. Training loss remains consistently lower and smoother than validation loss, which fluctuates more, indicating possible overfitting. | Line | Multiple Line |
Springer_3rd_066.png | This heatmap is titled 'Confusion Matrix'. The x-axis represents Predicted Label, and the y-axis represents True Label. Both the x- and y-axis have the same seven categories, including 'Pathogen', 'Object of hazard', 'Main symptoms', etc. The cell value represents the count of classification, ranging from 0 to 100. | The chart shows that the diagonal values are high, indicating correct classification for all categories. The variables 'Pathogen' and 'Main Symptoms' have 100 correct classification. | Heatmap | Heatmap |
Springer_2nd_1012 copy.png | This scatter plot with regression lines shows the value of mass loss on the y-axis from 0% to 90%, against the Fungai infection level on the x-axis from 1 to 5. Five levels of richness is listed in the legend: 1, 2, 4, 8, and 16. | The chart shows that the mass loss tends to increase with fungal infection level, especially at higher richness levels. Richness 16 shows the steepest positive trend. | Scatter | Scatter plot with regression line |
3465.png | This bar chart shows Screen Coverage for Different Navigation Aid. 4 types of navigation aid is listed on the x-axis, Screen coverage is shwon on the y-axis, ranging from 0.02 to 0.10. | No aid' shows highest screen coverage, around 0.05, 'compass' shows the lowest screen coverage, just 0.03. | Bar | Simple Bar |
springer_0522.png | This line chart shows the year on the x-axis from 2014 to 2022. The y-axis shows the "million people" from 7 to 10.5. | The chart shows a decrease at first but then a strong rise again in 2015 from 7.25 to 10 million. | Line | Simple Line |
0369.png | This multiple LIne chart shows Number of daily reviews on the y-axis in range of 0 to 3000, against Days on the x-axis. Each line present one conditon: Appstore, and Playstore. | For playstore, there is a large decline for number of daily reviews from 3000 to below at the begining, and the it showly decreases to 0 afterwards. For Appstore conditon, there is a small peak of number of daily reviewd on April 10, and then it slowly decreases to 0 as well. | Line | Multiple Line |
springer_4198.png | This bar chart shows Taskwise Recognition Accuracy across four tasks. The y-axis represents Taskwise Recognition Accuracy, ranging from 91 to 97. The x-axis lists four tasks: Right, Forward, Stop, and Left. The legend indicates one condition: Accuracy | The chart shows that "Right" and "Stop" tasks have higher accuracy scores, around 94.45. "Left" and "Forward" tasks have lower accuracy, around 92.22 and 93.34. | Bar | Simple Bar |
arXiv_0390.png | This scatter plot is titled "Phase Space Plot", it shows the relationship between Smoothed Node 6 Value on the y-axis in range of -0.01 to 0.03 against Smoothed Node 2 Value on the x-axis in range of -0.0768 to -0.0759. Three conditions are compared as indicated in the legend: Normal, Threshold and Anomaly. | The plot shows a cluster with "Normal" and "Treshold" condition on the right bottom corner of the plot with a narrow spread. The "Anomaly" condition has a much more wider spread, spreading from the middle to the left upper corner of the plot. | Scatter | Simple Scatter |
extra_0362.png | This pie chart is titled 'The compostion of IMI_RPF in the rice'. It shows the breakdown of seven categories: ACE, NIT, CLO, DIN, IMI, THI, and THIA. Each category is represented by one segment, with percentage labelled. | The chart shows that the THIA category takes up 51.2%, followed by the THI category, 34.1%. Other categories have less percentage. | Pie | Pie |
springer_4387.png | This horizontal bar chart shows the Top-30 Most Salient Terms. The y-axis lists the categories, and the x-axis shows the overall term frequency, ranging from 0 to 120.000. | The most frequent topic is "doctor," with around 110,000 occurrences, followed by "provides" with around 70.000." The topic with least frequency is "to help" with almost 0. | Bar | Simple Bar |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.