In recent years, the interest in recommendation systems (RSs) has dramatically increased, as they have become main components of all online stores. The aims of an RS can be multifaceted, related not only to the increase in sales or the convenience of the customer, but may include the promotion of alternative environmentally friendly products or to strengthen policies and campaigns. In addition to accurate suggestions, important aspects of contemporary RSs are therefore to align with the particular marketing goals of the e-shop and with the stances of the targeted audience, ensuring user acceptance, satisfaction, high impact, and achieving sustained usage by customers. The current review focuses on RS related to retail shopping, highlighting recent research efforts towards enhanced e-shops and more efficient sustainable digital marketing and personalized promotion. The reported research was categorized by main approach, key methods, and specialized e-commerce problems addressed, while technological aspects were linked with marketing aspects. The increasing number of papers in the field showed that it has become particularly popular, following the explosive growth in e-commerce and mobile shopping. The problems addressed have expanded beyond the performance of the core algorithms to the business aspects of recommendation, considering user acceptance and impact maximization techniques. Technologies have also shifted from the improvement of classic filtering techniques to complex deep learning architectures, in order to deal with issues such as contextualization, sequence-based methods, and automatic feature extraction from unstructured data. The upcoming goals seem to be even more intelligent recommendations that more precisely adapt not only to users’ explicit needs and hidden desires but also to their personality and sensitivity for more sustainable choices.

Research into session-based recommendation systems (SBSR) has attracted a lot of attention, but each study focuses on a specific class of methods. This work examines and evaluates a large range of methods, from simpler statistical co-occurrence methods to embeddings and SotA deep learning methods. This paper analyzes theoretical and practical issues in developing and evaluating methods for SBSR in e-commerce applications, where user profiles and purchase data do not exist. The major tasks of SBRS are reviewed and studied, namely: prediction of next-item, next-basket and purchase intent. For physical retail shopping where no information about the current session exists, we treat the previous baskets purchased by the user as previous sessions drawn from a loyalty system. Mobile application scenarios such as push notifications and calling tune recommendations are also presented. Recommender models using graphs, embeddings and deep learning methods are studied and evaluated in all SBRS tasks using different datasets. Our work contributes a number of very interesting findings. Among all tested models, LSTMs consistently outperform other methods of SBRS in all tasks. They can be applied directly because they do not need significant fine-tuning. Additionally, they naturally model the dynamic browsing that happens in e-commerce web applications. On the other hand, another important finding of our work is that graph-based methods can be a good compromise between effectiveness and efficiency. Another important conclusion is that a “temporal locality principle” holds, implying that more recent behavior is better suited for prediction. In order to evaluate these systems further in realistic environments, several session-based recommender methods were integrated into an e-shop and an A/B testing method was applied. The results of this A/B testing are in line with the experimental results, which represents another important contribution of this paper. Finally, important parameters such as efficiency, application of business rules, re-ranking issues, and the utilization of hybrid methods are also considered and tested, providing comprehensive useful insights into SBRS and facilitating the transferability of this research work to other domains and recommendation scenarios.

The problem of results merging in distributed information retrieval environments has been approached by two different directions in research. Estimation approaches attempt to calculate the relevance of the returned documents through ad-hoc methodologies (weighted score merging, regression etc) while download approaches, download all the documents locally, partially or completely, in order to estimate "first hand" their relevance. Both have their advantages and disadvantages. It is assumed that download algorithms are more effective but they are very expensive in terms of time and bandwidth. Estimation approaches on the other hand, usually rely on document relevance scores being returned by the remote collections in order to achieve maximum performance. In addition to that, regression algorithms, which have proved to be more effective than weighted scores merging, rely on a significant number of overlap documents in order to function effectively, practically requiring multiple interactions with the remote collections. The new algorithm that is introduced reconciles the above two approaches, combining their strengths, while minimizing their weaknesses. It is based on downloading a limited, selected number of documents from the remote collections and estimating the relevance of the rest through regression methodologies. The proposed algorithm is tested in a variety of settings and its performance is found to be better than estimation approaches, while approximating that of download.

Purpose: In federated search, a query is sent simultaneously to multiple resources and each one of them returns a list of results. These lists are merged into a single list using the results merging process. In this work, the authors apply machine learning methods for results merging in federated patent search. Even though several methods for results merging have been developed, none of them were tested on patent data nor considered several machine learning models. Thus, the authors experiment with state-of-the-art methods using patent data and they propose two new methods for results merging that use machine learning models.Design/methodology/approach: The methods are based on a centralized index containing samples of documents from all the remote resources, and they implement machine learning models to estimate comparable scores for the documents retrieved by different resources. The authors examine the new methods in cooperative and uncooperative settings where document scores from the remote search engines are available and not, respectively. In uncooperative environments, they propose two methods for assigning document scores.Findings: The effectiveness of the new results merging methods was measured against state-of-the-art models and found to be superior to them in many cases with significant improvements. The random forest model achieves the best results in comparison to all other models and presents new insights for the results merging problem.Originality/value: In this article the authors prove that machine learning models can substitute other standard methods and models that used for results merging for many years. Our methods outperformed state-of-the-art estimation methods for results merging, and they proved that they are more effective for federated patent search.

Conventional recommendation methods such as collaborative filtering cannot be applied when long-term user models are not available. In this paper, we propose two session-based recommendation methods for anonymous browsing in a generic e-commerce framework. We represent the data using a graph where items are connected to sessions and to each other based on the order of appearance or their co-occurrence. In the first approach, called Hierarchical Sequence Probability (HSP), recommendations are produced using the probabilities of items’ appearances on certain structures in the graph. Specifically, given a current item during a session, to create a list of recommended next items, we first compute the probabilities of all possible sequential triplets ending in each candidate’s next item, then of all candidate item pairs, and finally of the proposed item. In our second method, called Recurrent Item Co-occurrence (RIC), we generate the recommendation list based on a weighted score produced by a linear recurrent mechanism using the co-occurrence probabilities between the current item and all items. We compared our approaches with three state-of-the-art Graph Neural Network (GNN) models using four session-based datasets one of which contains data collected by us from a leather apparel e-shop. In terms of recommendation effectiveness, our methods compete favorably on a number of datasets while the time to generate the graph and produce the recommendations is significantly lower.