Recently, click-through rate (CTR) prediction models have evolved from shallow methods to deep neural networks. Most deep CTR models follow an Embedding\&MLP paradigm, that is, first mapping discrete id features, e.g. user visited items, into low dimensional vectors with an embedding module, then learn a multi-layer perception (MLP) to fit the target. In this way, embedding module performs as the representative learning and plays a key role in the model performance. However, in many real-world applications, deep CTR model often suffers from poor generalization performance, which is mostly due to the learning of embedding parameters. In this paper, we model user behavior using an interest delay model, study carefully the embedding mechanism, and obtain two important results: (i) We theoretically prove that small aggregation radius of embedding vectors of items which belongs to a same user interest domain will result in good generalization performance of deep CTR model. (ii) Following our theoretical analysis, we design a new embedding structure named res-embedding. In res-embedding module, embedding vector of each item is the sum of two components: (i) a central embedding vector calculated from an item-based interest graph (ii) a residual embedding vector with its scale to be relatively small. Empirical evaluation on several public datasets demonstrates the effectiveness of the proposed res-embedding structure, which brings significant improvement on the model performance.

While advancements in big data analytics have done a good job at helping marketers target mass markets and people of like interests, they fall short of understanding a person's unique interests and going that extra mile of treating people like individuals. While there is no argument that people have overlapping interests, there is a false assumption that just because a person falls into a certain category (i.e. A person will express hundreds of different interests that extend beyond any given category; this is what makes a person unique. Instead, they tend to bucket people into broad categories. Big data needs many data points to evaluate whether an interest is statistically significant.

Everyone says we're in the first inning of AI (or even the first at-bat) but what does that mean specifically in regards to how AI is being utilized and is understood by the leading tech companies? You would think Google would have a major AI advantage baked into its business model, and it does. Search is a data scientist and machine learning professional's dream. Your entire core (search) business is data. Google Search has data on a perhaps unprecedented scale, because of the volume of its users and the volume of searches.

Artificial Intelligence(AI) is gonna drive the world. In this process, network analysis is gonna play a big role. Slowly we are moving from traditional data analytics to the most charming data science in a way that we are moving from traditional marketing to digital marketing.. As a result of it, companies become smarter in catering to the needs of their customers, in predicting their sales volume, resource needs, the next problems, the right recommendations and in automating many manual processes. Moreover, so-far-piled up data suddenly become an asset for the companies for a reason the data science is gonna find the hidden insights and patterns.

Existing personalisation approaches, such as collaborative filtering or content based recommendations, are highly dependent on the domain and/or the source of the data. Therefore, there is a need for more accurate means to capture and model the interests of the user across domains, and to interlink them in a semantically-enhanced interest graph. We propose a new approach for multi-source, cross-genre recommendations that can exploit the heterogeneous nature of user profile data, which has been aggregated from multiple personalised web services, such as blogs, wikis and microblogs. Our approach is based on the Spreading Activation model that exploits intrinsic links between entities across a number of data sources. The proposed method is highly customizable and applicable both to generic and specific recommendation scenarios and use cases. With the growing number of Social Web applications in the enterprise (blogs, wikis, micro blogging, etc.), it becomes difficult for knowledge workers to avoid content overload and to quickly identify relevant people, communities and information. We demonstrate the application of our approach in an industrial use case that involves recommendation of social semantic data across multiple services in a distributed collaborative environment.