Methods for Mining and Summarizing Text Conversations by Giuseppe Carenini, Gabriel Murray, Raymond Ng

By Giuseppe Carenini, Gabriel Murray, Raymond Ng

A result of net Revolution, human conversational facts -- in written types -- are gathering at a good looking price. even as, advancements in speech expertise permit many spoken conversations to be transcribed. participants and enterprises have interaction in e mail exchanges, face-to-face conferences, running a blog, texting and different social media actions. The advances in typical language processing supply considerable possibilities for those "informal files" to be analyzed and mined, hence growing quite a few new and helpful functions. This publication provides a collection of computational the right way to extract info from conversational information, and to supply traditional language summaries of the knowledge. The e-book starts off with an outline of uncomplicated suggestions, comparable to the variations among extractive and abstractive summaries, and metrics for comparing the effectiveness of summarization and diverse extraction initiatives. It additionally describes many of the benchmark corpora utilized in the literature. The ebook introduces extraction and mining equipment for appearing subjectivity and sentiment detection, subject segmentation and modeling, and the extraction of conversational constitution. It additionally describes frameworks for undertaking discussion act attractiveness, determination and motion merchandise detection, and extraction of thread constitution. there's a particular specialize in acting these types of projects on conversational facts, resembling assembly transcripts (which exemplify synchronous conversations) and emails (which exemplify asynchronous conversations). Very contemporary methods to accommodate blogs, dialogue boards and microblogs (e.g., Twitter) also are mentioned. the second one half this ebook makes a speciality of normal language summarization of conversational facts. It supplies an outline of numerous extractive and abstractive summarizers constructed for emails, conferences, blogs and boards. It additionally describes makes an attempt for construction multi-modal summarizers. final yet no longer least, the booklet concludes with recommendations on issues for extra improvement. desk of Contents: advent / history: Corpora and review equipment / Mining textual content Conversations / Summarizing textual content Conversations / Conclusions / ultimate innovations

Show description

Read or Download Methods for Mining and Summarizing Text Conversations (Synthesis Lectures on Data Management) PDF

Best data mining books

Mining of Massive Datasets

The recognition of the internet and web trade presents many super huge datasets from which details might be gleaned through info mining. This booklet specializes in functional algorithms which were used to resolve key difficulties in facts mining and which might be used on even the biggest datasets. It starts with a dialogue of the map-reduce framework, an immense device for parallelizing algorithms immediately.

Twitter Data Analytics (SpringerBriefs in Computer Science)

This short offers tools for harnessing Twitter info to find recommendations to complicated inquiries. The short introduces the method of accumulating information via Twitter’s APIs and gives ideas for curating huge datasets. The textual content supplies examples of Twitter facts with real-world examples, the current demanding situations and complexities of creating visible analytic instruments, and the easiest techniques to handle those concerns.

Advances in Natural Language Processing: 9th International Conference on NLP, PolTAL 2014, Warsaw, Poland, September 17-19, 2014. Proceedings

This publication constitutes the refereed lawsuits of the ninth foreign convention on Advances in average Language Processing, PolTAL 2014, Warsaw, Poland, in September 2014. The 27 revised complete papers and 20 revised brief papers provided have been rigorously reviewed and chosen from eighty three submissions. The papers are prepared in topical sections on morphology, named entity attractiveness, time period extraction; lexical semantics; sentence point syntax, semantics, and computing device translation; discourse, coreference solution, automated summarization, and query answering; textual content category, info extraction and data retrieval; and speech processing, language modelling, and spell- and grammar-checking.

Analysis of Large and Complex Data

This publication bargains a picture of the state of the art in type on the interface among facts, machine technology and alertness fields. The contributions span a large spectrum, from theoretical advancements to sensible purposes; all of them percentage a powerful computational part. the themes addressed are from the subsequent fields: statistics and information research; desktop studying and data Discovery; info research in advertising; facts research in Finance and Economics; info research in drugs and the existence Sciences; information research within the Social, Behavioural, and future health Care Sciences; information research in Interdisciplinary domain names; type and topic Indexing in Library and data technology.

Extra resources for Methods for Mining and Summarizing Text Conversations (Synthesis Lectures on Data Management)

Example text

3. EVALUATION METRICS FOR SUMMARIZATION 33 • The group presented their final budget. The first sentence is from a system summary and the second sentence is from a gold-standard human summary of the document. We can see that the bigram final budget occurs in each sentence, so we say that there is a bi-gram overlap between this sentence and the gold-standard. If we permit intervening terms between the words of the bigram, we can identify further overlaps, which are called skip bigram overlaps. The following pair of sentences illustrates skip bigram overlap: • So let’s look at the final revised budget.

A major reason why the summarization community has been slow to adopt “official” evaluation metrics (compared with, say, the machine translation community) is precisely owing to conflicting results regarding such correlations in different domains. Liu and Liu [2010] is a recent example of work trying to measure the usefulness of a popular intrinsic evaluation software package (ROUGE, described in Chapter 2) on noisy conversational data. Extrinsic evaluations, on the other hand, measure the usefulness of a summary in aiding some real-world task, such as document classification or reading comprehension.

Machine summaries are then annotated for SCUs as well and can be scored based on the sum of SCU weights compared with the sum of SCU weights for an optimal summary. , SCU6 ), and two possible optimal summaries containing four SCUs are indicated. These summaries are optimal because they each contain all of the SCUs of weight 4, the highest weight level, and the remaining SCUs from weight 3, the next highest level of the Pyramid. Using the SCU annotation, one can calculate both precision-based and recall-based summary scores for a given machine summary.

Download PDF sample

Rated 4.94 of 5 – based on 25 votes