Get the element with the highest occurrence in an array

As per George Jempty's request to have the algorithm account for ties, I propose a modified version of Matthew Flaschen's algorithm.

function modeString(array) {
  if (array.length == 0) return null;

  var modeMap = {},
    maxEl = array[0],
    maxCount = 1;

  for (var i = 0; i < array.length; i++) {
    var el = array[i];

    if (modeMap[el] == null) modeMap[el] = 1;
    else modeMap[el]++;

    if (modeMap[el] > maxCount) {
      maxEl = el;
      maxCount = modeMap[el];
    } else if (modeMap[el] == maxCount) {
      maxEl += "&" + el;
      maxCount = modeMap[el];
    }
  }
  return maxEl;
}

This will now return a string with the mode element(s) delimited by a & symbol. When the result is received it can be split on that & element and you have your mode(s).

Another option would be to return an array of mode element(s) like so:

function modeArray(array) {
  if (array.length == 0) return null;
  var modeMap = {},
    maxCount = 1,
    modes = [];

  for (var i = 0; i < array.length; i++) {
    var el = array[i];

    if (modeMap[el] == null) modeMap[el] = 1;
    else modeMap[el]++;

    if (modeMap[el] > maxCount) {
      modes = [el];
      maxCount = modeMap[el];
    } else if (modeMap[el] == maxCount) {
      modes.push(el);
      maxCount = modeMap[el];
    }
  }
  return modes;
}

In the above example you would then be able to handle the result of the function as an array of modes.

I guess you have two approaches. Both of which have advantages.

Sort then Count or Loop through and use a hash table to do the counting for you.

The hashtable is nice because once you are done processing you also have all the distinct elements. If you had millions of items though, the hash table could end up using a lot of memory if the duplication rate is low. The sort, then count approach would have a much more controllable memory footprint.