You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
KNN right now does not maintain the state of the top k from one round of tweets to the next round. To see this in full effect have the tweets come in at a slow rate (set the sleep for 4000) and you will notice that for each query the one tweet record that is added is the top k every time. i.e. the query focal point and k are saved but the top k are not. I believe this is due to the built in priority queue that is holding the current top k needs to be maintained in a DStream. At the moment Thamir agreed to look into this but anyone is welcome help since I need to refocus my efforts on data mining for these next few weeks.
The text was updated successfully, but these errors were encountered:
KNN right now does not maintain the state of the top k from one round of tweets to the next round. To see this in full effect have the tweets come in at a slow rate (set the sleep for 4000) and you will notice that for each query the one tweet record that is added is the top k every time. i.e. the query focal point and k are saved but the top k are not. I believe this is due to the built in priority queue that is holding the current top k needs to be maintained in a DStream. At the moment Thamir agreed to look into this but anyone is welcome help since I need to refocus my efforts on data mining for these next few weeks.
The text was updated successfully, but these errors were encountered: