SMS based Interface for FAQ Retrieval
Kothari, Govind and Negi, Sumit and Faruquie, Tanveer A. and Chakaravarthy, Venkatesan T. and Subramaniam, L. Venkata

Article Structure

Abstract

Short Messaging Service (SMS) is popularly used to provide information access to people on the move.

Introduction

The number of mobile users is growing at an amazing rate.

Prior Work

There has been growing interest in providing access to applications, traditionally available on Internet, on mobile devices using SMS.

Problem Formulation

We View the input SMS S as a sequence of tokens S = 31, 32, .

Pruning Algorithm

We now describe algorithms for computing the maximum scoring question 62*.

System Implementation

In this section we describe the weight function, the preprocessing step and the creation of lists L1,L2,...,Ln.

Experiments

We validated the effectiveness and usability of our system by carrying out experiments on two FAQ data sets.

Conclusion

In recent times there has been a rise in SMS based QA services.

Topics

similarity measure

Appears in 15 sentences as: Similarity Measure (1) similarity measure (15)
In SMS based Interface for FAQ Retrieval
  1. For each term t in the dictionary and each SMS token 3,, we define a similarity measure a(t, 3,) that measures how closely the term 75 matches the SMS token 3,.
    Page 3, “Problem Formulation”
  2. Combining the similarity measure and the inverse document frequency (idf) of t in the corpus, we define a weight function to (t, 3,).
    Page 3, “Problem Formulation”
  3. The similarity measure and the weight function are discussed in detail in Section 5.1.
    Page 3, “Problem Formulation”
  4. The weight function is a combination of similarity measure between t and Si and Inverse Document Frequency (idf) of t. The next two subsections explain the calculation of the similarity measure and the idf in detail.
    Page 4, “System Implementation”
  5. 5.1.1 Similarity Measure
    Page 4, “System Implementation”
  6. For term t E D and token 3%- of the SMS, the similarity measure a(t, 81) between them is
    Page 4, “System Implementation”
  7. the Levenshtein distance between them is less, the similarity measure defined in Equation 2 will be high.
    Page 5, “System Implementation”
  8. We explain the rationale behind using the EditDistanceSMs in the similarity measure 0405,31) through an example.
    Page 5, “System Implementation”
  9. As a result the similarity measure between “gud” and “good” will be higher than that of “gud” and “guided”.
    Page 5, “System Implementation”
  10. Combining the similarity measure and the idf of t in the corpus, we define the weight function w(t, 81') as
    Page 5, “System Implementation”
  11. We prefer terms that have high similarity measure i.e.
    Page 5, “System Implementation”

See all papers in Proc. ACL 2009 that mention similarity measure.

See all papers in Proc. ACL that mention similarity measure.

Back to top.

question answering

Appears in 8 sentences as: Question Answering (1) question answering (7)
In SMS based Interface for FAQ Retrieval
  1. This has resulted in the growth of SMS based Question Answering (QA) services.
    Page 1, “Abstract”
  2. In this work we present an automatic FAQ-based question answering system for SMS users.
    Page 1, “Abstract”
  3. Most of these contact center based services and other regular services like “AQA 63336”1 by Issuebits Ltd, GTIP2 by AlienPant Ltd., “Tex-perts”3 by Number UK Ltd. and “ChaCha”4 use human agents to understand the SMS text and re-spond.U)these ShdS quefies.’The naune oftex-ting language, which often as a rule rather than exception, has misspellings, nonstandard abbreviations, transliterations, phonetic substitutions and omissions, makes it difficult to build automated question answering systems around SMS technology.
    Page 1, “Introduction”
  4. Unlike other automatic question answering systems that focus on generating or searching answers, in a FAQ database the question and answers are already provided by an expert.
    Page 1, “Introduction”
  5. In this paper we present a FAQ-based question answering system over a SMS interface.
    Page 1, “Introduction”
  6. The information retrieval based system treat question answering as an information retrieval problem.
    Page 2, “Prior Work”
  7. In FAQ based question answering , where FAQ provide a ready made database of question-answer, the main task is to find the closest matching question to retrieve the relevant answer (Sneiders, 1999) (Song et al., 2007).
    Page 2, “Prior Work”
  8. We address the challenges in building a FAQ-based question answering system over a SMS interface.
    Page 2, “Prior Work”

See all papers in Proc. ACL 2009 that mention question answering.

See all papers in Proc. ACL that mention question answering.

Back to top.

natural language

Appears in 3 sentences as: Natural language (1) natural language (2)
In SMS based Interface for FAQ Retrieval
  1. Some businesses have recently allowed users to formulate queries in natural language using SMS.
    Page 1, “Introduction”
  2. These systems generally adopt one of the following three approaches: Human intervention based, Information Retrieval based, or Natural language processing based.
    Page 2, “Prior Work”
  3. The natural language processing based system tries to fully parse a question to discover semantic structure and then apply logic to formulate the answer (Molla et al., 2003).
    Page 2, “Prior Work”

See all papers in Proc. ACL 2009 that mention natural language.

See all papers in Proc. ACL that mention natural language.

Back to top.

scoring function

Appears in 3 sentences as: scoring function (3)
In SMS based Interface for FAQ Retrieval
  1. the retrieved questions is formalized using a scoring function .
    Page 2, “Introduction”
  2. Based on the weight function, we define a scoring function for assigning a score to each question in the corpus Q.
    Page 3, “Problem Formulation”
  3. For each token 3,, the scoring function chooses the term from Q haVing the maximum weight; then the weight of the n chosen terms are summed up to get the score.
    Page 3, “Problem Formulation”

See all papers in Proc. ACL 2009 that mention scoring function.

See all papers in Proc. ACL that mention scoring function.

Back to top.

semantically similar

Appears in 3 sentences as: Semantically similar (1) semantically similar (2)
In SMS based Interface for FAQ Retrieval
  1. The Pruning algorithm uses this dictionary to retrieve semantically similar questions.
    Page 6, “System Implementation”
  2. To retrieve answers for SMS queries that are semantically similar but lexically different from questions in the FAQ corpus we use the Synonym dictionary described in Section 5.2.
    Page 6, “System Implementation”
  3. Figure 4: Semantically similar SMS and questions
    Page 6, “System Implementation”

See all papers in Proc. ACL 2009 that mention semantically similar.

See all papers in Proc. ACL that mention semantically similar.

Back to top.

PDF available from www.aclweb.org.
Send us an email