Analytics for Services

Voice of Customer Analytics is an end-to-end managed service solution to provide actionable findings for customer retention and extension. Heterogeneous structured and unstructured data sources like Customer Relationship Management (CRM) information, customer surveys, call transcripts, agent logs, and activity records are tapped and linked together. The solution combines text analytics with more traditional data mining processes to deliver actionable insights based on information that is inherent in customer interactions. Analytics techniques in the Voice of Customer Analytics transform customer utterances into a meaningful set of usable data that can be categorized, searched, and analyzed. The research team works with the services teams on the underlying enabling technology and a host of new research challenges.

Researchers: Indrajit Bhattacharya, Shantanu Godbole, Ajay Gupta, Ashish Verma

Hiring and staffing of skilled resources is an important activity in large IT services organizations. Such organizations receive hundreds of job applications every day and their hiring organization faces the tremendous problem of matching those applications against various job openings and short-listing a small percentage of highly skilled candidates for interview. The goal of this project is to develop a decision support tool using text mining, information extraction, and optimization techniques to help recruiters identify the most suitable candidates easily. Various challenges involved in such tasks include error-free extraction of skills from resumes written in non-standard formats, inferring skill level of a person from 'skill mentions', matching and choosing best skill levels for various job requirements, etc. IRL Researchers work closely with the HR organization of IBM to develop this technology.

Researchers: Nandakishore Kambhatla, Vijil Chenthamarakshan, Swati Challa

Sensei is a web-enabled tool to evaluate a person's spoken English skills. It uses advanced speech processing techniques to evaluate spoken English skills under various parameters, including pronunciation, grammar, syllable stress and comprehension. Sensei enables people to improve their English speaking skills via the Internet and provides a score that can be used to determine their skill levels in spoken English. A highly cost effective solution, Sensei can help organizations, especially those in customer services or knowledge space, improve the language skills of their employees.

Researchers: Ashish Verma, Himanshu Chauhan, Kundan Kandhway, Om Deshmukh

(Web Access by Voice & Easy Web Browsing)

Web Access by Voice (WAV) is a system to allow access to information on World Wide Web (WWW) by using voice, instead of traditional browsing. WAV provides relevant information from Internet, given a user's query over phone. The idea behind WAV is to decouple Web information from Web browsing. Given a user's query, the system identifies the relevant Web page or an interaction on the Web site; gathers required inputs for transaction from the user (through voice) and relays back the resulting information. WAV does not impose any additional constraint either on the existing Web structure or HTML page syntax. Moreover, to achieve similar results as manual browsing this system leverages the Web browser interface itself, thus proving effective in accessing Web-page content which is generated by JavaScript, AJAX and other similar features.

Easy Web Browser (EWB) is a Web browser created for visually impaired (partially or totally) people to access information on the Internet easily. It provides a user-friendly Web interface, including character enlargement, reading of text aloud and optimization of Web pages by changing font sizes and background colors according to user preferences. EWB is a cost-effective solution as it can be hosted on the Web site by the service provider and end users don't need to buy it.

Researchers: Ashish Verma, Himanshu Chauhan, Kundan Kandhway, Om Deshmukh, Vijay Garg

In large IT services organizations, efficient matching of practitioners to project openings is critical to profitability. Practitioners and project openings are characterized by structured attributes such as job role, skill set, location city and unstructured attributes such as job description and resumes. Complexity arises due to inter-relationships between attributes (for example the hierarchical relationship between skills), difficulty in extracting the level of experience from resumes, missing or incorrect data, as well as aligning the process to customer requirements. Workforce Optimization is an interdisciplinary project that uses constrained optimization, machine learning, and text analytics techniques to develop a decision support solution which not only provides effective matching of practitioners to projects but also reduces overhead involved in information sharing by end users. In developing this solution, IRL researchers drawn from multiple backgrounds will work closely with end users as well as other IBM research labs.

Researchers: : Nandakishore Kambhatla, Rohit Lotlikar, Munish Goyal, Debapriyo Majumdar, Gyana Parija, Pranav Gupta, Sambuddha Roy, Kashyap Dixit, Soujanya Soni

Customer satisfaction is key to retain and grow client base in retail banking. One of the primary components of customer satisfaction is wait time -- the time spent by a customer in the bank before getting serviced. Retail banks are loosing market share in the absence of smart systems to reduce the customer wait time. This is particularly true for the developing countries where a branch is the primary channel for banking and the average wait time could be extremely high. In this work, we developed a system for scheduling transactions to reduce the average wait time of customers while providing differentiated Quality of Service (QoS) based on customer profile and transaction characteristics. Our system has been deployed in one of the leading retail banks in India for a trial period of one month. The overall wait time reduced by 29 percent, priority customers experienced reduction of 57 percent in their wait times and the wait time of customers in the most important category decreased by 80 percent.

Researchers: Sameep Mehta, Gyana Parija, Vikas Kedia

The broad area of interest in this collaboration is "service supply chains". We are focusing on mathematical modeling techniques that will enable service delivery organizations to improve their way of functioning by leveraging cooperation among different people of the organization. The project explores the possibility of significant efficiency improvements in service delivery by analyzing data available on people/agents interaction within the organization. As a case-study, we propose to consider the "orchestration model" of service delivery where a composite service is created by outsourcing several of its components to a subset of a pool of independent service providers distributed geographically. The knowledge about various aspects of the network of service providers is essential in achieving efficiency in creating a composite service. The goal of the collaboration is to investigate if efficiency of composite service delivery can be improved by leveraging and inferring from data on pattern of interaction and the cost thereof, among the independent service providers.

Researchers:

• (IRL) Vinayaka Pandit, Sameep Mehta, Kashayp Dixit
• (ISB Hyderabad) S. Kameshwaran

Each year, wildfires cause massive destructions across the globe. According to the U.S. National Interagency Fire Center, the level of damages caused by wildfires has increased significantly over the last few years. Since 2003, IBM has been working with the U.S. Department of Agriculture Forest Service to control and manage this natural disaster. Fire Program Analysis (FPA), started under IBM's On Demand Innovation Services, uses unique mathematical algorithms to determine where wildfires are likely to occur, and helps optimize government funds and resources for battling those fires. FPA provides the National Wildfire Coordination Group - five federal agencies including the Forest Service, Bureau of Land Management, National Park Service, Fish and Wildlife Service, and Bureau of Indian Affairs - with a single, integrated system to support wildfire preparedness analysis, planning, and budgeting. The system shows the best choice of tools, and the number of acres that can be protected, under different budget scenarios.

Researchers

• (IRL) Gyana Parija, Soujanya Soni, Pranav Gupta
• (Watson Research Center) Tarun Kumar

Rapidly growing service delivery organizations need efficient solutions to manage their business and reduce the cost of growth. These delivery businesses run round the clock as the support customers in different time zones. The cost of adding new physical infrastructure and space to tap incoming business is a key driver of total cost for growth. Physical space availability and its effective management is the key infrastructure attribute for the business delivery apart from human resources. Hence a sizeable impact can be realized if space utilization can be increased by optimal allocation around different work shifts. Optimal space utilization can bring down the seats required to serve the existing business and also will allow them to commit to new business needs with the existing capacity. IRL researchers are developing the technology to meet various strategic and tactical business objectives in this setting.

Researchers: Gyana Parija, Pranav Gupta, Jayanta Basak, Kashyap Dixit