Based on recent advances in natural language modeling and those in text generation capabilities, we propose a novel data augmentation method for text classification tasks. We use a powerful pre-trained neural network model to artificially synthesize new labeled data for supervised learning. We mainly focus on cases with scarce labeled data. Our method, referred to as language-model-based data augmentation (LAMBADA), involves fine-tuning a state-of-the-art language generator to a specific task through an initial training phase on the existing (usually small) labeled data. Using the fine-tuned model and given a class label, new sentences for the class are generated. Our process then filters these new sentences by using a classifier trained on the original data. In a series of experiments, we show that LAMBADA improves classifiers' performance on a variety of datasets. Moreover, LAMBADA significantly improves upon the state-of-the-art techniques for data augmentation, specifically those applicable to text classification tasks with little data.

Social media platforms such as blogs, wikis and file sharing have become very popular in enterprises. Despite their effectiveness in increasing collaboration in the organization, employees are overloaded with information originating from these many sources and find it hard to orient themselves in the stream of events occurring in their organizational news feed. In this paper we identify what makes an event in an organizational social media platform important to employees. Once important factors of an event to an employee are identified, the stream of events can be personalized and prioritized based on those and thus reduce the overload and assist in work efficiency. Through interviews and two extensive user surveys, the first hypothetical and the second empirical, we identified which factors of an event make it important and compare results from the hypothetical and empirical surveys.

The 'like' button, introduced by Facebook several years ago, has become one of the most prominent icons of social media. Similarly to other popular social media features on the web, enterprises have also recently adopted it. In this paper, we present a first comprehensive study of liking activity in the enterprise. We studied the logs of an enterprise social media platform within a large global organization along a period of seven months, in which 393,720 'likes' were performed. In addition, we conducted a survey of 571 users of the platform's 'like' button. Our evaluation combines quantitative and qualitative analysis to inspect what employees like, why they use the 'like' button, and to whom they give their 'likes'.

We study personalized item recommendation within an enterprise social media application suite that includes blogs, bookmarks, communities, wikis, and shared files. Recommendations are based on two of the core elements of social media - people and tags. Relationship information among people, tags, and items, is collected and aggregated across different sources within the enterprise. Based on these aggregated relationships, the system recommends items related to people and tags that are related to the user. Each recommended item is accompanied by an explanation that includes the people and tags that led to its recommendation, as well as their relationships with the user and the item. We evaluated our recommender system through an extensive user study. Results show a significantly better interest ratio for the tag-based recommender than for the people-based recommender, and an even better performance for a combined recommender. Tags applied on the user by other people are found to be highly effective in representing that user's topics of interest.

This work investigates personalized social search based on the user's social relations -- search results are re-ranked according to their relations with individuals in the user's social network. We study the effectiveness of several social network types for personalization: (1) Familiarity-based network of people related to the user through explicit familiarity connection; (2) Similarity-based network of people "similar" to the user as reflected by their social activity; (3) Overall network that provides both relationship types. For comparison we also experiment with Topic-based personalization that is based on the user's related terms, aggregated from several social applications. We evaluate the contribution of the different personalization strategies by an off-line study and by a user survey within our organization. In the off-line study we apply bookmark-based evaluation, suggested recently, that exploits data gathered from a social bookmarking system to evaluate personalized retrieval. In the on-line study we analyze the feedback of 240 employees exposed to the alternative personalization approaches. Our main results show that both in the off-line study and in the user survey social network based personalization significantly outperforms non-personalized social search. Additionally, as reflected by the user survey, all three SN-based strategies significantly outperform the Topic-based strategy.

We study personalized recommendation of social software items, including bookmarked web-pages, blog entries, and communities. We focus on recommendations that are derived from the user's social network. Social network information is collected and aggregated across different data sources within our organization. At the core of our research is a comparison between recommendations that are based on the user's familiarity network and his/her similarity network. We also examine the effect of adding explanations to each recommended item that show related people and their relationship to the user and to the item. Evaluation, based on an extensive user survey with 290 participants and a field study including 90 users, indicates superiority of the familiarity network as a basis for recommendations. In addition, an important instant effect of explanations is found - interest rate in recommended items increases when explanations are provided.

This work investigates cluster labeling enhancement by utilizing Wikipedia, the free on-line encyclopedia. We describe a general framework for cluster labeling that extracts candidate labels from Wikipedia in addition to important terms that are extracted directly from the text. The "labeling quality" of each candidate is then evaluated by several independent judges and the top evaluated candidates are recommended for labeling.

No abstract available.

Users tend to store huge amounts of files, of various formats, on their personal computers. As a result, finding a specific, desired file within the file system is a challenging task. This article addresses thedesktop searchproblem by considering various techniques for ranking results of a search query over the file system. First, basic ranking techniques, which are based on various file features (e.g., file name, access date, file size, etc.), are considered and their effectiveness is empirically analyzed. Next, two learning-based ranking schemes are presented, and are shown to be significantly more effective than the basic ranking methods. Finally, a novel ranking technique, based on query selectiveness, is considered for use during the cold-start period of the system. This method is also shown to be empirically effective, even though it does not involve any learning.

This paper addresses the desktop search problem by considering varioustechniques for ranking results of a search query over thefile system. First, basic ranking techniques, which are based ona single file feature (e.g., file name, file content, access date, etc.)are considered. Next, two learning-based ranking schemes are presented, and are shown to be significantly more effective than the basic ranking methods. Finally, a novel ranking technique, based on query selectiveness is considered,for use during the cold-start period of the system. This method isalso shown to be empirically effective, even though it does notinvolve any learning.