Research

Semantic Network Analysis in Social Sciences

We are increasingly exposed to, produce and share information through traditional and social media channels. Text is one of the most valuable artifacts, providing scholars with endless opportunities to study our society. In many of my recent studies I use semantic network analysis to understand the biases of large texts. This method enables to transform any given text into a visual map of words co-occurring together. In this recent book, together with colleagues from all over the world, we offer some fundamental guidelines for researchers with no previous knowledge in the field, and present various recent applications of semantic network analysis to study news, political speeches, social media, user-generated content, and even to help researchers organize interview transcripts and conduct meta-analysis of literature reviews.

To read more on those studies click here

International News Flow Online: Understanding the world through news site mining

Why are some countries more newsworthy than others? What are the similarities and differences in the scope of international news presented in different languages and cultures? How does international news affect our perception of the world? My studies together with Menahem Blondhein from the Hebrew University in Jerusalem aim at exploring the bias of international news their global and local patterns.

I compare the international scope of online newspapers, news portals, and news aggregators in different languages and cultures, using innovative web mining techniques and network analysis. The book "International News Flow Online: Global Views with Local Perspectives" explores the theory of news flow around the world, and analyses many of its dimensions such as the global standing of the United States, the Middle Eastern conflicts as seen around the world, and, the effect of financial news. In doing so, it unveils new patterns, meanings and implications of international news on our perception of the world.

To read more on those studies click here

Provocation in International News

Going deeper into the content of international news and the framing of countries, two studies conducted together with Sandrine Boudana. One is theoretical about the use of provocation narratives in international news, and the other is empirical. We found that North Korea is by far the most frequently mentioned country in the context of provocation, and that the actions of North Korea are never directly explained or justified. Provocations, we claim, is tiny yet powerful word that tells us the meta-narrative of western news, and helps to justify the geopolitical ambitions of the U.S. in east Asia.

To read more on those studies click here

Information search and the digital divide

Rather than a simple division between those who have and those who have-not (access to the Internet), the second level digital divide is seen as the different information uses, search strategies and skills among online users. For example, certain information searches (i.e., for news, tax, law, government, society, or business) can provide economic and socio-political advantages. They can help online users to find jobs, compare prices of products and services, establish online business, acquire an education, increase their social and political involvement, and so on. The book "Google and the Digital Divide: The Bias of Online Knowledge" deals with the relationship between online search, search engines and the digital divide.

In this line of research I am interested in developing new methods and analytical tools to study what information is being sought in different countries and to what extent this information is politically and economically related. My studies are based on longitudinal analyses of the search queries in popular search engines and include a large number of countries. Among others, I examine the content, heterogeneity and accuracy of searches in order to shed new light on the digital divide. My PhD student, Nathan Stolero, is studying how young and adult users differ in their information seeking behavior.

To read more on this study click here

Seeking science information online

These series of studies were conducted in collaboration with Ayelet Baram-Tsabari from the Technion Institute. We aim at exploring the scientific interests and needs of online users worldwide. We have been employing data from Google Trends in order to identify public interest in science, and understand the potential and limitations of these publicly available tools for scientific research. You can read our blog post on when and why people Google Nobel Prize winners and their discoveries.

To read more on this study click here

A systematic procedure for detecting news biases: The case of Israel in European news sites

In this project, funded by the SNF, we develop a systematic and structured procedure for sentiment analysis in the news, as well as a database of keywords that incorporate positive and negative opinions toward Israel in different languages. We mine 14 large newspapers in five countries—the UK, France, Germany, Italy, and Switzerland—over a period of six months and rate their respective orientations. Our findings clearly show that news reports are largely critical and negative toward Israel, with British news being the most critical, Italian news the most sensational, and German, French and Swiss news relatively more neutral. Opinions featured in the news are not in line with public opinion as presented in annual surveys of each country.

To read more on this study click here

Tracing the DNA of cultural products: The families and network structure of internet memes

Understanding cultures has been perhaps one of my most important research goals, and it appears in one way or another in almost all of my studies. One such an attempt was carried out together with Limor Shiffman, who has studied extensively the field of globalization, humor, and internet memes. Together with Asaf Nissenbaum and Nathan Stolero we tried to understand what are the cues that unite internet memes to be in the same family, and how they are related to the rest of the mainstream popular culture. We therefore designed a large codebook to analyze the content, form and communication structure of about 1,000 instances of 20 large internet meme families.

The most exciting discovery for us was the significant relationship between the uniqueness and cohesiveness of meme families. The more unique a meme family is - the more it is cohesive. Uniqueness was measured in terms of the proximity of the content, form and communication of a meme family from all the rest of the popular memes. Cohesiveness was measured as the extent of similarity among the instances of the same meme family. This seems to be a pretty obvious idea that unique groups are also more homogeneous and cohesive when it comes to the formation of social groups. Another surprising discovery was that meme families that are largely based on a picture or an object are usually more unique and cohesive as a meme family. It reminds a bit of the power of symbols such as a flag to unite people together. Abstract ideas, on the other hand, such as democracy or monotheism usually calls for the formation of much more heterogeneous groups.

To read more on this study click here