Context. Long and consistent sunspot area records are important for understanding long-term solar activity and variability. Multiple observatories around the globe have regularly recorded sunspot areas, but such individual records only cover restricted periods of time. Furthermore, there are systematic differences between these records and require cross-calibration before they can reliably be used for further studies. Aims. We produce a cross-calibrated and homogeneous record of total daily sunspot areas, both projected and corrected, covering the period between 1874 and 2019. In addition, we generated a catalog of calibrated individual group areas for the same period. Methods. We compared the data from nine archives: Royal Greenwich Observatory (RGO), Kislovodsk, Pulkovo, Debrecen, Kodaikanal, Solar Optical Observing Network (SOON), Rome, Catania, and Yunnan Observatories, covering the period between 1874 and 2019. Cross-comparisons of the individual records were done to produce homogeneous and inter-calibrated records of daily projected and corrected areas. As in earlier studies, the basis of the composite is formed by the data from RGO. After 1976, the only datasets used are those from Kislovodsk, Pulkovo, and Debrecen observatories. This choice was made based on the temporal coverage and the quality of the data. While there are still 776 days missing in the final composite, these remaining gaps could not be filled with data from the other archives as the missing days lie either before 1922 or after 2016 and none of the additional archives cover these periods. Results. In contrast to the SOON data used in previous area composites for the post-RGO period, the properties of the data from Kislovodsk and Pulkovo are very similar to those from the RGO series. They also directly overlap the RGO data in time, which makes their cross-calibration with RGO much more reliable. Indeed, comparing our area catalog with previous such composites, we find improvements both in data quality and coverage. We also computed the daily Photometric Sunspot Index, which is widely used, for example, in empirical reconstructions of solar irradiance.

Complementary recommendations play an important role in surfacing the relevant items to the customers. In the cross-selling scenario, some customers might present more exploratory shopping behaviors and prefer more diverse complements, while other customers show less exploratory (or more conventional) shopping behaviors and want to have a deep dive of less diverse types of complements. The existence of two distinct shopping behaviors reflects users' different shopping intents and requires complementary recommendations to be adaptable based on the user's shopping intent. Although many studies focus on improving the recommendations through post-processing techniques, such as user-item-level personalized ranking and diversification of recommendations, they fail to address such a requirement. First, many user-item-level personalization methods cannot explicitly model the preference of users in two types of shopping behaviors and their intent on the corresponding complementary recommendations. Second, most of the diversification methods increase the heterogeneity of the recommendations. However, users' intent on conventional complementary shopping requires more homogeneity of the recommendations, which is not explicitly modeled. The present study tries attempts to solve these problems by the personalized diversification strategies for complementary recommendations. To address the requirement of modeling heterogenized and homogenized complementary recommendations, we propose two diversification strategies, heterogenization and homogenization, to re-rank complementary recommendations based on the determinantal point process (DPP). We use transaction history to estimate users' intent on more exploratory or more conventional complementary shopping. With the estimated user intent scores and two diversification strategies, we propose an algorithm to personalize the diversification strategies dynamically. We demonstrate the effectiveness of our re-ranking algorithm on the publicly available Instacart dataset.

No abstract