0.8592
Result of MCRS Item–Based CF
Top 10 Main Aspects Extracted
| Configuration | Dataset | Yelp | Place, food, service, restaurant, price menu, staff, drink, and lunch | ||||
|---|---|---|---|---|---|---|---|
| #asp. | Sub-asp | Yelp | TripAdvisor | Amazon | |||
| 10 | Y | 0.864 | 0.8245 | 0.811 | TripAdvisor | Hotel, room, staff, location, service, breakfast, restaurant, bathroom, price, view | |
| 10 | N | 0.8643 | 0.8252 | 0.8117 | |||
| 50 | Y | 0.8641 | 0.8254 | 0.8118 | Amazon | game, graphic, story, character, player price, gameplay, controller level, and music | |
| 50 | N | 0.8648 | 0.826 | 0.8124 |
It is noted that, to get better efficiency, we need to configure all the datasets in top 10 neighbors.
In the next step, they compared the algorithm with matrix factorization (MF) algorithms for getting better baselines. Besides, it is specified that TripAdvisor dataset has an unchangeable set of six features. For every analysis like “Cleanliness”, “location”, “value”, “service”, “sleep quality”, and “overall”. They have differentiated their way with a MCRS algorithm that depends on those aspects. In Table 3.5, we can see that their method gets the better of against all the baselines. These results surely established that their perception, since they proved that their approach could overpower both strategies exploiting single ratings and MCRS algorithms.
When multiple-criteria user-to-user CF is used as recommender algorithm, then the best overall results are obtained [5].
3.4.2 User Preference Learning in Multi-Criteria Recommendation Using Stacked Autoencoders by Tallapally et al.
Here, they come up with a stacked autoencoders which is a DNN approach to use the multiple-criteria ratings. They implemented a model which is configured to analyze the connection in the middle of every client’s criteria and general rating efficiency. Test outcomes are based on practical datasets like Yahoo! Movies dataset and TripAdvisor dataset. It illustrates that this approach can perform both single-criteria systems and multi-criteria approaches on different performance matrix [4].
Now, if we look on their proposed performance evaluation and result analysis, then it will be cleared that how much efficiency this model can achieve.
3.4.2.1 Dataset and Evaluation Matrix
In this paper, they have used two datasets based on real world from tourism and movie domains that are used to evaluate the performance. They hold on to the sample data of the users who reviewed at least five hotels and hotels that were reviewed by at least five users to obtain working data subset from TA.
They used subset that carry more than 19,000 rating instances by more than 3,100 users with around 3,500 hotels which has a high sparsity of 99.8272%. In addition, YM data are generated as shown in Tables 3.3 to 3.5. For analyzing the performance of this method, they used Mean Absolute Error (MAE) which is known for its simplicity, accuracy, and popularity [4].
Table 3.3 Dataset.
Result = YM 10-10
Result = YM 20-20
| Technique | MAE | GIMAE | GPIMAE | F1 | Technique | MAE | GIMAE | GPIMAE | F1 | |
|---|---|---|---|---|---|---|---|---|---|---|
| MF [10] | 0.8478 | 0.7461 | 0.6765 | 0.5998 | MF [10] | 0.7397 |
|
