Studying, developing, and experimenting contextual advertising systems

The World Wide Web has grown so fast in the last decade and it is today a vital daily part of people. The Internet is used for many purposes by an ever growing number of users, mostly for daily activities, tasks, and services. To face the needs of users, an efficient and effective access to information is required. To deal with this task, the adoption of Information Retrieval and Information Filtering techniques is continuously growing. Information Re-trieval (IR) is the field concerned with searching for documents, information within documents, and metadata about documents, as well as searching for structured storage, relational databases, and the World Wide Web. Infor- mation Filtering deals with the problem of selecting relevant information for a given user, according to her/his preferences and interest. Nowadays, Web advertising is one of the major sources of income for a large number of websites. Its main goal is to suggest products and services to the still ever growing population of Internet users. Web advertising is aimed at suggesting products and services to the users. A significant part of Web ad-vertising consists of textual ads, the ubiquitous short text messages usually marked as sponsored links. There are two primary channels for distributing ads: Sponsored Search (or Paid Search Advertising) and Contextual Ad-vertising (or Content Match). Sponsored Search advertising is the task of displaying ads on the page returned from a Web search engine following a query. Contextual Advertising (CA) displays ads within the content of a generic, third party, webpage. In this thesis I study, develop, and evaluated novel solutions in the field of Contextual Advertising. In particular, I studied and developed novel text summarization techniques, I adopted a novel semantic approach, I studied and adopted collaborative approaches, I started a conjunct study of Contex-tual Advertising and Geo-Localization, and I study the task of advertising in the field of Multi-Modal Aggregation. The thesis is organized as follows. In Chapter 1, we briefly describe the main aspects of Information Retrieval. Following, the Chapter 2 shows the problem of Contextual Advertising and describes the main contributes of the literature. Chapter 3 sketches a typical adopted approach and the eval-uation metrics of a Contextual Advertising system. Chapter 4 is related to the syntactic aspects, and its focus is on text summarization. In Chapter 5 the semantic aspects are taken into account, and a novel approach based on ConceptNet is proposed. Chapter 6 proposes a novel view of CA by the adoption of a collaborative filtering approach. Chapter 7 shows a prelim-inary study of Geo Location, performed in collaboration with the Yahoo! Research center in Barcelona. The target is to study several techniques of suggesting localized advertising in the field of mobile applications and search engines. In Chapter 8 is shown a joint work with the RAI Centre for Research and Technological Innovation. The main goal is to study and propose a system of advertising for Multimodal Aggregation data. Chapter 9 ends this work with conclusions and future directions.