deletions | additions       diff --git a/Personal reflections.md b/Personal reflections.md  index 96fbc49..a257dc8 100644  --- a/Personal reflections.md  +++ b/Personal reflections.md 
...
#Personal reflections  Stochastic The stochastic  nature of human behavior makes it difficult to operate with global patterns  This research patterns, such as the occupancy schedule standards. Reality is much more complex than that and with the rise of cheap electronics integrated into more and more smart buildings, occupancy monitoring and subsequent prediction for energy consumption optimization is a vital component of building energy portfolio analysis. OSL  makes it possible to predict occupancy patterns of buildings of similar scope but unfitted with sensor sensors (oftentimes a certain array of sensors is deployed in a location, then removed - such as the ones at Masdar Institute)  or having fewer or lower quality or reliability sensors  The above step is crucial for the implementation of sensors. I think  the smart grid and renewable energy systems requiring larger energy conservation for expensive storage  Not OSL is in no way  meant to replace but to complement agent-based models – opening the way to incorporate feedback processes in the modeling  C4.5 is modeling. C4.5, the algorithm OSL uses for creating  a decision tree model is  suitablealgorithm  for learning the occupancy presence in offices: offices and it leads to a claimed  90.3 % accuracy. 4 occupancy profiles identified using the k-means algorithm with Davies-Bouldin optimization  Day-to-day variation in profiles observed for the same offices  Large variation but recurring patterns activity times lead However this leads  to developing working user profiles some of potential problems of the paper.  I think that MTD OSL  is a powerful concept with a very elegant algorithm. The paper is well-written arguably split into two somewhat disconnected parts: the decision tree model  and it the occupancy profile creation through clustering. It  is easy to follow for anyone with a background not clear  in data mining - for me it represented a bit of a difficulty what ways are the occupancy profiles (the mined or the known ones?) used  as I have a base in statistics and network science but not this kind inputs for the clustering algorithm. Moreover, OSL also provides very little information about the methodology  oftechnical data mining that  the paper preconditions. It is important to put this paper in clustering - despite the fact  the right context: it time series clustering  is a theory paper, large and potentially very complex area \cite{Denton_2005}. There is no accuracy measure or reproducible, universal metric provided for  the latest in a long succession assessment  of topic modeling papers published by the author group over the past 4-5 years, which were sequentially built upon each other. First, they extensively studied the LDA algorithm, then they extended it with must-link mining the results  andeventually with cannot-link mining. Unless one goes through  the previous bibliography, then mathematical theory could be too fast-paced to follow. The description is quite clear until probable cross-validation process of  the cannot-links. I think decision tree model is not described either.   These flaws, together with a large and watery introduction decrease  the strongest weakness credibility  of the paper comes at this point, while they provide paper, despite  the mathematical model to map fact that  the cannot-links, they only explain authors of OSL have multiple well-received publications in  the method and its plausibility vaguely. building modeling field.