Recommender system

[1][4] Typically, the suggestions refer to various decision-making processes, such as what product to purchase, what music to listen to, or what online news to read.

It utilizes user metadata in order to discover and recommend appropriate content, whilst reducing ongoing maintenance and development costs.

[13] The differences between collaborative and content-based filtering can be demonstrated by comparing two early music recommender systems, Last.fm and Pandora Radio.

[15][16][17][18][19][20] Whereas Pandora needs very little information to start, it is far more limited in scope (for example, it can only make recommendations that are similar to the original seed).

Her idea was to create a system that asks users specific questions and classifies them into classes of preferences, or "stereotypes", depending on their answers.

Another early recommender system, called a "digital bookshelf", was described in a 1990 technical report by Jussi Karlgren at Columbia University, [34] and implemented at scale and worked through in technical reports and publications from 1994 onwards by Jussi Karlgren, then at SICS,[35][36] and research groups led by Pattie Maes at MIT,[37] Will Hill at Bellcore,[38] and Paul Resnick, also at MIT,[39][4] whose work with GroupLens was awarded the 2010 ACM Software Systems Award.

The weights denote the importance of each feature to the user and can be computed from individually rated content vectors using a variety of techniques.

Most instances of session-based recommender systems rely on the sequence of recent interactions within a session without requiring any additional details (historical, demographic) of the user.

Techniques for session-based recommendations are mainly based on generative sequential models such as recurrent neural networks,[63][66] transformers,[67] and other deep-learning-based approaches.

It is important to consider the risk of upsetting the user by pushing recommendations in certain circumstances, for instance, during a professional meeting, early morning, or late at night.

One example of a mobile recommender system are the approaches taken by companies such as Uber and Lyft to generate driving routes for taxi drivers in a city.

[76] This system uses GPS data of the routes that taxi drivers take while working, which includes location (latitude and longitude), time stamps, and operational status (with or without passengers).

In one method, known as HSTU (Hierarchical Sequential Transduction Units),[78] high-cardinality, non-stationary, and streaming datasets are efficiently processed as sequences, enabling the model to learn from trillions of parameters and to handle user action histories orders of magnitude longer than before.

By turning all of the system’s varied data into a single stream of tokens and using a custom self-attention approach instead of traditional neural network layers, generative recommenders make the model much simpler and less memory-hungry.

Offline evaluations are based on historic data, e.g. a dataset that contains information about how users previously rated movies.

In the context of recommender systems a 2019 paper surveyed a small number of hand-picked publications applying deep learning or neural methods to the top-k recommendation problem, published in top conferences (SIGIR, KDD, WWW, RecSys, IJCAI), has shown that on average less than 40% of articles could be reproduced by the authors of the survey, with as little as 14% in some conferences.

The articles considers a number of potential problems in today's research scholarship and suggests improved scientific practices in that area.

By 2011, Ekstrand, Konstan, et al. criticized that "it is currently difficult to reproduce and extend recommender systems research results," and that evaluations are "not handled consistently".

Said and Bellogín conducted a study of papers published in the field, as well as benchmarked some of the most popular frameworks for recommendation and found large inconsistencies in results, even when the same algorithms and data sets were used.

They conclude that seven actions are necessary to improve the current situation:[117] "(1) survey other research fields and learn from them, (2) find a common understanding of reproducibility, (3) identify and understand the determinants that affect reproducibility, (4) conduct more comprehensive experiments (5) modernize publication practices, (6) foster the development and use of recommendation frameworks, and (7) establish best-practice guidelines for recommender-systems research."

The AI-based recommender can analyze complex data sets, learning from user behavior, preferences, and interactions to generate highly accurate and personalized content or product suggestions.

Traditional methods often relied on inflexible algorithms that could suggest items based on general user trends or apparent similarities in content.

[121] These advanced methods enhance system capabilities to predict user preferences and deliver personalized content more accurately.

The ideas are as follows: An artificial neural network (ANN), is a deep learning model structure which aims to mimic a human brain.

Unlike regular machine learning where the underlying theoretical components are formal and rigid, the collaborative effects of neurons are not entirely clear, but modern experiments has shown the predictive power of ANN.

[121] Following are some examples: The Two-Tower model is a neural architecture[124] commonly employed in large-scale recommendation systems, particularly for candidate retrieval tasks.

[125] It consists of two neural networks: The outputs of the two towers are fixed-length embeddings that represent users and items in a shared vector space.

This model is highly efficient for large datasets as embeddings can be pre-computed for items, allowing rapid retrieval during inference.

Google Scholar provides an 'Updates' tool that suggests articles by using a statistical model that takes a researchers' authorized paper and citations as input.

Therefore, there is a risk that the market could become fragmented, leaving it to the viewer to visit various locations and find what they want to watch in a way that is time-consuming and complicated for them.