refine and improve the recommendation approach considering people with common interests; users evaluation of BROAD-RSI recommendations; evaluation of users' interaction in the BROAD-RSI through their logs. .
Use of standardized measures, including for patient characteristics associated with guideline recommendations, evaluation of guideline adherence and longer term patient outcomes will also be important.
The use of cost-benefit reviews for recommendation evaluations is also highlighted.
Yet, the suitability of a recommendation algorithm to support these use cases cannot be comprehensively evaluated by any recommendation evaluation measures proposed so far.
In this paper, we propose a method to expand the repertoire of existing recommendation evaluation techniques with a method to evaluate the discoverability and navigability of recommendation algorithms.
We have presented a novel evaluation method that expands the repertoire of recommendation evaluation measures with a technique to evaluate discoverability and navigability.
The parameters settings used for the semi-supervised pairwise recommendation evaluation are the same used in [9].
c Path B. To illustrate how the recommendation evaluation was performed for a complete workflow, consider again that someone wants to build the workflow shown in Fig. 11 b.
The application of established techniques from network science allows us to present a novel method that extends common evaluation measures towards a path-based evaluation and expands the arsenal of existing recommendation evaluation techniques with two dimensions that have not received sufficient attention so far.
A de-briefing on the preliminary findings of the evaluation (Preliminary Draft Report) 6. Presentation of main findings, conclusions and recommendations (Draft Evaluation Report) 7. A Final Evaluation Report The evaluation was designed as a summative, using non-experimental approach.
