[
  {
    "id": 0,
    "page": 12,
    "bounding_box": [
      133.72277323404947,
      56.89202880859375,
      461.55311414930554,
      175.64703369140625
    ],
    "latex_content": "\\begin{table}[ht]\n    \\centering\n    \\begin{tabular}{|l|l|l|}\n\\hline Step & Operations & Storage (not including storage of $A$) \\\\\n        \\hline  \\ref{alg2:init} & $O(mn)$ & $O(n)$ \\\\\n        \\hline  \\ref{alg2:max_arg}& $O(ns-s^2)$ & $O(1)$ \\\\\n        \\hline  \\ref{alg2:under_diag}& $O(s^2)$ & $O(s^2)$ \\\\\n        \\hline  \\ref{alg2:above_diag1}& $O(m(ns-s^2))$ & $O(m)$\\\\\n        \\hline  \\ref{alg2:above_diag2}& $O(s(ns-s^2))$ & $O(ns-s^2)$\\\\\n        \\hline  \\ref{alg2:above_diag3}& $O(ns-s^2)$ & $O(1)$\\\\\n        \\hline Total: & $O(mns) $ & $ O(ns) $\\\\ \\hline\n    \\end{tabular}\n    \\caption{Computational and storage complexities of Algorithm~\\ref{alg: q-less}.}\n\\end{table}",
    "extracted_content": [
      [
        "Step",
        "Operations",
        "Storage (not including storage of A)"
      ],
      [
        "1",
        "O(mn)",
        "O(n)"
      ],
      [
        "4",
        "O(ns s2)\n\u2212",
        "O(1)"
      ],
      [
        "7",
        "O(s2)",
        "O(s2)"
      ],
      [
        "10",
        "O(m(ns s2))\n\u2212",
        "O(m)"
      ],
      [
        "11",
        "O(s(ns s2))\n\u2212",
        "O(ns s2)\n\u2212"
      ],
      [
        "12",
        "O(ns s2)\n\u2212",
        "O(1)"
      ],
      [
        "Total:",
        "O(mns)",
        "O(ns)"
      ]
    ],
    "similarity_score": 0.62,
    "table_image": "images/1607.03456v1/table_0.png",
    "page_image": "pages/1607.03456v1/page_12.png"
  },
  {
    "id": 1,
    "page": 22,
    "bounding_box": [
      146.58177947998047,
      56.89202880859375,
      448.69320024762834,
      246.67999267578125
    ],
    "latex_content": "\\begin{table}[H]\n\\begin{center}\n    \\begin{tabular}{ |c|l|c|c| }\n    \\hline\n        $\\mu$    &  & ICPQR-based DM   & $\\mu$IDM~\\cite{Salhov:uIDM2014} \\\\\n    \\hline\n        \n    &  Dictionary size $s$            & $1,246$     & $2,305$      \\\\\n$0.1$    &  Execution time          & $43$ sec.      & $7$ hours  \\\\\n    &  Actual distortion           & $0.01$     & $0.03$   \\\\\n    \\hline\n   \n    &  Dictionary size $s$              & $752$      & $1,293$     \\\\\n$1$     &  Execution time          & $27$ sec.       & $71$ minutes  \\\\\n    &  Actual distortion           & $0.23$     & $0.61$   \\\\\n    \\hline\n      \n    &  Dictionary size  $s$             & $382$      & $630$       \\\\\n$5$    &  Execution time          & $17$ sec.      & $15$ minutes  \\\\\n    &  Actual distortion           & $3.91$     & $4.46$   \\\\\n    \\hline\n       \n    &  Dictionary size   $s$            & $190$      & $284$       \\\\\n$10$    &  Execution time          & $9$ sec.       & $4$ minutes  \\\\\n    &  Actual distortion           & $12.81$    & $13.24$  \\\\\n    \\hline\n    \\end{tabular}\n\\caption{Comparison between ICPQR-based DM and $\\mu$IDM algorithms related to dictionary size, execution time and actual distortion\\protect\\footnotemark, w.r.t. $\\timel{\\map\\Psi}{1}$, for various distortion parameters. Clearly, the actual distortion is bounded by $2\\mu$. Execution times are averaged over $10$ runs of the algorithms.} \\label{tbl:compare_icpqr_idm}\n\\end{center}\n\\end{table}",
    "extracted_content": [
      [
        "\u00b5",
        "",
        "ICPQR-based DM",
        "\u00b5IDM [41]"
      ],
      [
        "0.1",
        "Dictionary size s\nExecution time\nActual distortion",
        "1, 246\n43 sec.\n0.01",
        "2, 305\n7 hours\n0.03"
      ],
      [
        "1",
        "Dictionary size s\nExecution time\nActual distortion",
        "752\n27 sec.\n0.23",
        "1, 293\n71 minutes\n0.61"
      ],
      [
        "5",
        "Dictionary size s\nExecution time\nActual distortion",
        "382\n17 sec.\n3.91",
        "630\n15 minutes\n4.46"
      ],
      [
        "10",
        "Dictionary size s\nExecution time\nActual distortion",
        "190\n9 sec.\n12.81",
        "284\n4 minutes\n13.24"
      ]
    ],
    "similarity_score": 0.3744454303460515,
    "table_image": "images/1607.03456v1/table_1.png",
    "page_image": "pages/1607.03456v1/page_22.png"
  },
  {
    "id": 2,
    "page": 23,
    "bounding_box": [
      124.92357417515346,
      361.218017578125,
      470.35142735072543,
      450.2840270996094
    ],
    "latex_content": "\\begin{table}[ht]\n    \\centering\n    \\begin{tabular}{|c|c|c|c|c|}\n        \\hline Set & Size & \\# of anomalies & Accuracy [\\%] & False Alarms [\\%] \\\\\n        \\hline $\\set T_{mon}$ & $1,321$ & $1$ & $100$     & $0.68$\\\\\n        \\hline $\\set T_{tue}$ & $1,140$ & $53$ &$100$    & $0.53$\\\\\n        \\hline $\\set T_{wed}$ & $1,321$ & $16$ &$100$  & $0.08$\\\\\n        \\hline $\\set T_{thu}$ & $1,320$ & $24$ &$96$    & $1.74$\\\\\n        \\hline $\\set T_{fri}$ & $1,320$ & $18$ &$100$     & $0.15$\\\\\n        \\hline\n    \\end{tabular}\n    \\caption{Anomaly detection performances.}\n\\end{table}",
    "extracted_content": [
      [
        "Set",
        "Size",
        "# of anomalies",
        "Accuracy [%]",
        "False Alarms [%]"
      ],
      [
        "Tmon",
        "1, 321",
        "1",
        "100",
        "0.68"
      ],
      [
        "Ttue",
        "1, 140",
        "53",
        "100",
        "0.53"
      ],
      [
        "Twed",
        "1, 321",
        "16",
        "100",
        "0.08"
      ],
      [
        "Tthu",
        "1, 320",
        "24",
        "96",
        "1.74"
      ],
      [
        "Tfri",
        "1, 320",
        "18",
        "100",
        "0.15"
      ]
    ],
    "similarity_score": 0.8475452196382429,
    "table_image": "images/1607.03456v1/table_2.png",
    "page_image": "pages/1607.03456v1/page_23.png"
  }
]