Algorithm 4: Estimating Data Truthfulness

[Service Provider receives a data submission $d$] //Evaluate data truthfulness for each category of measurement. foreach $c$ in $d$  //Check if the measurement value for this category is within the limits set by the service provider.  if $m_c$ < $m_{c_{\min }}$ or $m_c$ > $m_{c_{\max }}$ then   $v$ = false //If not within the limits, set the validation status to false and exit the algorithm.   return $v$  end if  Read data $\left[d_c\right]$ for $c$ //Read the data for this category of measurement.  Get $h$ from $\left[d_c\right]$’s $PDF$ //Compute the height of the dataset’s PDF.  $LLF$ = $h-sum\left(\log \left(h\right)\right)$ //Compute the Log Likelihood Function.  //Use initial estimates for the mean and standard deviation.  Use initial estimates ${\mathsf{\mu }}_e$ and ${\mathsf{\sigma }}_e$  //Use MLE to estimate the two parameters (mean and standard deviation) for which the normal  //distribution best describes the data.  Compute $\mathsf{\mu }$ and $\mathsf{\sigma }$ by applying $MLE$ using $LLF$, ${\mathsf{\mu }}_e$ and ${\mathsf{\sigma }}_e$  //Read the standard deviation factor for this measurement category, compute the scaled standard  //deviation setting and set the threshold limits.  Read $f_{\mathsf{\sigma }}$  ${\mathsf{\sigma }}_{\mathrm{scaled}}$ = $\mathsf{\sigma }*f_{\mathsf{\sigma }}$  $l_{\min }= \mathsf{\mu }$ − ${\mathsf{\sigma }}_{\mathrm{scaled}}$  $l_{\max }= \mathsf{\mu }$ + ${\mathsf{\sigma }}_{\mathrm{scaled}}$  //Check if the measurement value is within the threshold limits.  if $m_c$ < $l_{\min }$ or $m_c$ > $l_{\max }$ then   $v$ = false//If not within the threshold limits, set the validation status to false and exit the algorithm.   return $v$  end if end foreach $v$ = true//If the loop has finished without exiting the algorithm, the data submission is a valid one.