Discovery of implicit and explicit connections between people using email utterance

Abstract

Abstract. This paper is about finding explicit and implicit connections between people by mining semantic associations from their email communications. Following from a sociocognitive stance, we propose a model called HALe which automatically derives dimensional representations of words in a high dimensional context space from an email corpus. These dimensional representations are used to discover a network of people based on a seed contextual description. Such a network represents useful connections between people not easily achievable by 'normal' retrieval means. Implicit connections are "lifted" by applying latent semantic analysis to the high dimensional context space. The discovery techniques are applied to a substantial corpus of real-life email utterance drawn from a small-to-medium size information technology organization. The techniques are computationally tractable, and evidence is presented that suggests appropriate explicit connections are being brought to light, as well as interesting, and perhaps serendipitous implicit connections. The ultimate goal of such techniques is to bring to light contextsensitive, ephemeral, and often hidden relationships between people, and between people and information, which pervade the enterprise.

Uncovering deep user context from blogs

“…Stop words (non-italics) were removed. [25] revealed that for the purposes of this research, it was not useful to preserve word order information, so the HAL vector of a word was represented by the addition of its row and column vectors. The quality of HAL vectors is influenced by the window size; the longer the window, the higher the chance of representing spurious associations between terms.…”

“…Previous work [25][26][27] has shown that HAL can be used successfully in extracting knowledge directly from email utterances, both explicit and tacit. The studies have also shown the validity of using a (modified) global association model like HAL, originally developed on a 300 million word corpus from Usenet, on datasets of a few sentences [26] to a few thousand emails or documents [25].…”

“…The studies have also shown the validity of using a (modified) global association model like HAL, originally developed on a 300 million word corpus from Usenet, on datasets of a few sentences [26] to a few thousand emails or documents [25]. For all of these reasons, HAL was chosen in [24] and here as the semantic space model by which a representation of "sense-of-self" could be captured from the discourse in blog entries.…”

Beyond Keyword Filtering for Message and Conversation Detection

“…Others have used patterns in email to determine group structure, e.g. [11], or to determine topics of mutual interest [10].…”

Building Connections among Loosely Coupled Groups: Hebb?s Rule at Work

“…For those reasons, and because most of the subjects we interviewed indicated that they use e-mail as a primary communication medium with others in their group we chose to sense interests from e-mail. Furthermore, e-mail has been shown to be an effective means of discovering shared interests [30,16]. We used interviews and a pilot deployment to determine the most salient shared interests and report on our findings below.…”